f.

age-«wt "39'

3

1.
9. 3

11
x. .1: . .3. 3.3.24...
ad...“ h;‘Jn-.41x»un. ‘ L)
V a .: Invru
. :s
:12... _. :3
. 9

RPM... 3.2..
9.35“. an
a p

.1:

. .E 2.
. $.25: a.
: . .
H.314. .fi 1

.. s

9...,

“3:1... . :Lflurgnurmp
:1 ih!.,.§..:.... ..
ha a... x .

3:: .
...I..,.&:Kn

.L.v»

Lu. :91 an.

.1 It:
:- n:!3!tl.
a 3:35. :3...
5.12.531.»

3...}. t
15:31...

 

 

L..D .izi .
.v? 1 :

 

HOG

Min? WillWIWWWI

3 1293 01388 2331

 

 

This is to certify that the
dissertation entitled

A stump Prairie Landscape in Northern Michigan.
Soils, Forest Vegetation, Logging, and Fire

presented by

Linda R. Barrett

has been accepted towards fulfillment
of the requirements for

Ph-D- degree in GeogmpL

Wax

Major professor

 

Date 10/12/95

 

MS U it an Afl'trmative Action/Equal Opportunity Institution 0-1277!

  
 
 

 
 

LIBRARY
Michigan State
University

 
    

PLACE N RETURN BOX to remove We checkout from your record.
To AVOID FINES return on or before dete due.

DATE DUE DATE DUE DATE DUE
2 2007

MAGIC2
11999

W '
”we:
"Maul”!

MSUI-Anf‘“ A ._.. - --_r A -'

 

 

 

 

 

 

 

 

 

 

 

 

 

E

 

 

 

 

 

 

 

      
 
 
   
   
  
     
   

. _I‘

g" ".fi-A STUMP PRAIRIE LANDSCAPE IN NORTHERN MICHIGAN:
- SOILS, FOREST VEGETATION, LOGGING, AND FIRE

By

Linda R. Barrett

A DISSERTATION

1 Submitted to r . . .l I.
> ; Michigan State University '
1’ Inp'a'rtiai fulfillment of the requirem' ents ' ‘ ’ 4‘
Rum; gigtrr r - for the degree 0f - - x. 4 j ‘_

 

   

‘ ‘T .
"LLK'L‘ ‘ '..~ ,' ~

 

[pocrron OF PHILOSOPHY

442i «4.4?11"

ABSTRACT

A STUMP PRAIRIE LANDSCAPE IN NORTHERN MICHIGAN:
SOILS, FOREST VEGETATION, LOGGING, AND FIRE

By

Linda R. Barrett

Parts of a sandy plain located in northern Michigan today are “stump prairie,”
devoid of trees, although prior to the logging and fires of the late 19th century they
supported dense forest. Nearby, otherwise similar sites have regenerated to forest. The
aims of this study were to determine how site and human disturbance patterns are related
to patterns of long-standing changes in this ecosystem and to examine the possible impact
of these forest regeneration patterns on active soil processes.

Evidence from the General Land Office survey notes suggests that original forest
composition has a strong relationship to forest regeneration patterns, possibly due to the
manner in which early logging was accomplished. White pine (Pinus strobus) was
especially prevalent in the pre-logging forest of current stump prairie areas, but sugar
maple (Acer saccharum) was nearly lacking there. Land ownership and tax records
suggest that stump prairie sites were acquired and logged at least as early as adjacent
forested sites. Tree rings and recent stumps provide evidence that in currently forested
areas loggers left more trees to provide shade and seed sources for forest recovery.

Soil parent material texture is not significantly different between the forest and
stump prairie sites. Spodic horizon development, as shown by degree of ortstein
cementation and Fe, Al, and organic matter contents, is slightly stronger where forest

regeneration has occurred than in areas that have remained stump prairie. Most of the

 

extractable Fe and Al is present in organically-bound forms, but inorganic forms become
more important in the lower B horizon. Ortstein content represents the primary
difference between the soils in forested and stump prairie areas.

In order to study current soil development processes, bags of cation exchange and
chelating resins were buried in forested and unforested sites. Although variability within
and between pedons was very high, slightly greater amounts of Fe and A1 were sorbed in
B horizons of soils of forested than of unforested sites, suggesting that podzolization
processes are more active in the forest than in the stump prairie. These data highlight the

importance of forest vegetation in maintaining the spodic (Bs) horizon.

ACKNOWLEDGEMENTS

I thank first of all my advisor, Dr. Randall Schaetzl, for the untiring support he
has given me during the course of my doctoral program. His advice and confidence in
my abilities have been invaluable. The other members of my doctoral committee have
also been generous with advice and encouragement: Dr. Delbert Mokma, Dr. David
Lusch, Dr. Michael Velbel, and Dr. Jay Harman.

I also thank Dr. Walter Loope of the National Biological Survey (formerly of
Pictured Rocks National Lakeshore) for his encouragement and ideas in the early stages
of this project. Mr. Bruce Veneberg of the Michigan Department of Natural Resources
answered questions about the Kingston Plains area. Peter Scull and Bob Hunckler
provided field assistance. Johan Liebens was a source of encouragement as he also
worked to finish his dissertation.

This dissertation would have been impossible without the support of my husband,
Bob. He was a cheerful unpaid field assistant, helped in tree and plant identification, got
photocopies from the archives downtown, and took over unaccustomed household duties
during the last stressful push to finish. My children, Isaac and Aaron, were patient
through long days of field work, and understanding as they saw less and less of their
mother.

Financial support was provided by a Doctoral Dissertation Research Improvement
Grant fi'om the National Science Foundation (grant number SBR—9405356), and also by a
grant from the Starkey Fund of the Association of American Geographers. The
Department of Geography at Michigan State University gave a Graduate Office
Fellowship for field work in 1995 and provided the use of field equipment. Laboratory
facilities for analysis of soil samples were provided by Dr. Delbert Mokma of the

Department of Crop and Soil Sciences. Access to historical aerial photographs was
contributed by the Center for Remote Sensing at Michigan State University and by the
Michigan Department of Natural Resources. The Center for Remote Sensing also

furnished imagery interpretation and computer facilities.

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LIST OF TABLES viii
LIST OF FIGURES 3:
INTRODUCTION AND STATEMENT OF PROBLEM ............................. 1
Structure of the dissertation 2
LITERATURE REVIEW 1
Stump Prairies 3
Possible origins 4

Role of human disturbance vs. environmental factors ................ 6

Pine forests, fire, and succession in northern Michigan ......................... 8

Fire regimes, forest types, and soil 3.. . ‘ . ‘ 10

Active soil processes and soil degradation 1]
Nineteenth century logging in the Upper Peninsula ............................. 13
ogging , " 13
Logging-related fires 14

STUDY ARE A 17
General 17
Geomorphology 1 7

Soils 7O
Vegetation and lumbering 71
METHODS 22
Current forest patterns 77

Stand , 29

Land cover/stand boundaries 75

Factor 1: Pre—disturbance forest patterns 75

Factor 2: Substrate and soil pmperties ‘4’)

Field ‘L J ‘4?

Laboratory methods 34

In situ monitoring of soil processes ‘45

Factor 3: Logging era fires and logging r 16
RESULTS AND DISCUSSION ‘7
Current forest patterns ‘47

Stand composition 27

Land cover/ stand boundaries 44

 

vi

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Factor 1: Pre-disturbance forest patterns 54
Surveyor line irs 54

Witness tree locations 62

Witness tree diameters 71

Stumps 78

Comparisons with current stand (‘nmpnsitinn Rs

Summary 88

Factor 2: Substrate and soil properties 91
Substrate, soil, and current forest patterns ................................ 91

Parent material properties and forest patterns 91

Properties of soil development and forest patterns 100

Transects 107

Soil development processes 1 12

Clay, silt, and pH 112

Fe, Al, and 0C 117

In situ monitoring of soil processes 136

Summary 142

Factor 3: Logging era fires and logging practices 143
Logging 143

Fires: Evidence for bunting 153

SUMMARY AND CONCLUSIONS 159
Suggestions for future ‘ 163
APPENDIX A: DATA TABLES 165
APPENDIX B: PEDON DESCRIPTIONS 186
LIST OF REFERENCES 199

 

vii

 

LIST OF TABLES

Table 1. Common and scientific names associated with surveyors’

 

 

 

prairie stands .

 

 

 

 

 

 

pedons.

pedons.

 

 

notations of tree specres. 77
Table 2. Diameter at breast height (DBH) and importance values by tree
species for forest and stump prairie areas. 28
Table 3. Importance values of tree species for the nine sampled forest
stands . 39
Table 4. Importanlce values of tree species for the eight sampled stump
' ‘ 41
Table 5. Absolute density of trees in the 17 sampled stands. 42
Table 6. Citation frequency of the tree species in surveyors’ line
descriptions for three current landcover classifications. 55
Table 7. Frequencies and standardized residuals expressing the degree of
association between species and current landcover classification. ............................ 63
Table 8. Relative density, relative dominance, and modified importance
values of witness trees grouped by current landuse classification. ............................ 72
Table 9. Diameter at breast height (DBH)1 and absolute density for
logging era stumps in sampled stands Rn
Table 10. Importance values and ranks of tree species as compiled from
GLO notes and from modern sampled stands. 86
Table 11. Uppermost C horizon particle size distributions for sampled
92
Table 12. Banding and lamellae encountered below the solum in sampled
94
Table 13. Solum weighted particle size distributions for sampled pedons. ...................... 95
Table 14. Ratios between coarse and fine sand ’r " 99
Table 15. B horizon weighted means for selected soil properties. ................................. 101
Table 16. Solum weighted means for selected soil properties. ....................................... 102
Table 17. Means of B51 ortstein and B51 matrix subsamples for selected
soil properties compared for forest and stump prairie soils. .................................... 103
Table 18. Selected mean soil properties of le ortstein and le matrix
subsamples for all the sampled pedons. 104

 

viii

 

Table 19. Summary of selected mean soil property relationships with
regard to current landcover type and soil ortstein status.

120

 

Table 20. Mean amounts of cations sorbed on cation exchange and
chelating resins in forest and stump prairie pedons.

139

 

Table 21. Representative ages from ring counts of selected large trees in

156

 

sampled stands.

Table 22. Particle size analysis data from the sampled pedons. ........................

Table 23. Chemical data from the sampled pedons.

............. 165

171

 

Table 24. Fe and Al ratio data from the sampled pedons.

177

 

Table 25. Cations sorbed on cation exchange resins in the sampled
pedons.

182

 

Table 26. Cations sorbed on chelating resins in the sampled pedons. ...............

............. 184

LIST OF FIGURES

Figure 1. Map of the study area, Alger County, Michigan. Hatching
represents forested areas; unhatched polygons are current stump
prairie. Black areas represent water bodies. Scale 1:62,500. ................................... 18

Figure 2. Map of the six township study area, Alger and Schoolcraft
Counties, Michigan. Inset square denotes area of field study area
shown in Figure 1. Black areas represent water bodies. Scale
1:154,000. 19

 

Figure 3. Location of sampled stands in the study area. See Figure 1 for
reference. 7‘;

 

Figure 4. Map of simplified current landcover classes for six townships in
Alger County, Michigan as of 1981. Inset square denotes area of
field study area. 79

Figure 5. Map of section lines with at least three timber species listed by
the surveyor and with a length of > 2700 fi. (830 m; > 1/2 mile) afier
the line-in-polygon overlay. Hatching represents current (1981)
stump prairie areas. 30

Figure 6. Location of soil color and ortstein content transects within the
study area. 2‘;

Figure 7. Aerial photographs of the study area, sections 29 and 30 of T 48
N, R 15 W. 1) 1939; 2) 1986 45

Figure 8. Aerial photographs of the study area, sections 8, 9, l6, and 17 of
T48N,R15 W. l)1939; 2) 1986. 46

Figure 9. Aerial photographs of the study area, sections 10 and 11 of T 48
N, R15 W. 1) 1939; 2) 1986 47

Figure 10. Aerial photographs of the study area, sections 27 and 28 of T
48 N,R 15W. 1) 1939; 2) 1986 50

 

Figure 11. Aerial photographs of the study area, sections 8 and 17 of T 48
N, R 15 W. 1) 1939;2)1986 51

Figure 12. Aerial photographs of the study area, sections 20 and 21 of T
48N,R15 W. 1) 1939; 2) 1986 S7

 

Figure 13. Survey lines for which the surveyor listed sugar maple (solid
lines), beech (short dashes), or hemlock (long dashes) as the most
abundant timber type. Hatching represents present stump prairie
areas. S6

 

Figure 14. Survey lines for which the surveyor listed white pine as the
most abundant timber type. Hatching represents present stump
prairie areas.

 

 

Hatching represents present stump prairie areas.

Figure 16. Survey lines for which the surveyor described the soil as “2d
rate”. Hatching represents present upland forest areas.

 

Figure 17. Distribution of witness trees by species. Hatching represents

Yellow birch witness trees. C) Beech witness trees

 

Figure 18. Distribution of witness trees by species. Hatching represents

 

pine witness trees. C) Jack pine witness trees
Figure 19. Distribution of witness trees by species. Hatching represents

witness trees. C) Spruce witness trees.

 

Figure 20. Distribution of witness trees by species. Hatching represents

Hemlock witness trees.

 

Figure 21. Distribution of red maple witness trees. Hatching represents
present stump prairie areas

 

Figure 22. Diameter size class distributions from GLO notes for white

currently mapped as stump prairie.

 

Figure 23. Diameter size class distributions from GLO notes for beech
trees in areas currently mapped as upland forest.

 

Figure 24. Diameter size class distributions from GLO notes for sugar
maple trees in areas currently mapped as upland forest

 

57
Figure 15. Survey lines for which surveyor described the soil as “3d rate”.

59

60
present upland forest areas. A) Sugar maple witness trees. B)

64
present stump prairie areas. A) White pine witness trees. B) Red

65
present swamp areas. A) Tamarack witness trees. B) White cedar

67
present stump prairie areas. A) Balsam fir witness trees. B)

62

70
pine trees in A) areas currently mapped as upland forest and B) areas

74

75

77

79

Figure 25. A representative view of stumps on the stump prairie. .........................

Figure 26. Average stump diameters by location of sampled stands

Figure 27. Diameter size class distributions for stumps in A) forest stands
and B) stump prairie stands

Figure 28. Percent medium sand by depth for four representative pedons.
Error bars signify high and low values in horizons where ortstein was
sampled separately.

Figure 29. Ratio of the coarse sand fraction to the fine sand fraction by
depth for four representative pedons. Error bars signify high and low
values in horizons where ortstein was sampled separately.

27

84

97

98

Figure 30. Percent of sample points in which ortstein was detected for
each quadrat on forest/stump prairie transects.

 

Figure 31. Arithmetic mean of recorded Munsell hue for the sample
points in each quadrat on forest/stump prairie transects.

 

Figure 32. Arithmetic mean of recorded Munsell value for the sample
points in each quadrat on forest/stump prairie transects.

 

Figure 33. Arithmetic mean of recorded Munsell chroma for the sample
points in each quadrat on forest/stump prairie transects.

 

Figure 34. Percent clay by depth for four representative pedons. Error
bars signify high and low values in horizons where ortstein was
sampled separately.

 

Figure 35. Percent silt by depth for four representative pedons. Error bars
signify high and low values in horizons where ortstein was sampled
separately.

 

Figure 36. pH by depth for four representative pedons. Error bars signify
high and low values in horizons where ortstein was sampled
separately.

 

Figure 37. OC content by depth for four representative pedons. Error bars
signify high and low values in horizons where ortstein was sampled
separately.

 

Figure 38. ODOE values by depth for four representative pedons. Error
bars signify high and low values in horizons where ortstein was
sampled separately.

 

Figure 39. Extractable Fe content by depth for four representative pedons. ..................
Figure 40. Extractable Al content by depth for four representative pedons. ..................
Figure 41. Si0 content by depth for four representative pedons. Error bars

signify high and low values in horizons where ortstein was sampled
separately.

 

Figure 42. Fee/Fed values by depth for four representative pedons. Error
bars signify high and low values in horizons where ortstein was
sampled separately.

 

Figure 43. (Feo - Fe )/F e values by depth for four representative pedons.
Error bars signify high and low values in horizons where ortstein was
sampled separately.

 

Figure 44. (A10 - A1 M)/A1 values by depth for four representative pedons.
Error bars sigmfy high and low values in horizons where ortstein was
sampled separately.

 

xii

108

109

110

111

113

115

116

118

119

121

122

124

125

127

128

Figure 45. Inorganic amorphous A1 (A10 - Alp) content by depth for four
representative pedons. Error bars signify high and low values in
horizons where ortstein was sampled separately.

 

Figure 46. (A10 - Alp)/Si0 values by depth for four representative pedons.
Error bars signify high and low values in horizons where ortstein was
sampled separately.

 

Figure 47. Fep/Alp values by depth for four representative pedons. Error
bars sigmfy high and low values in horizons where ortstein was
sampled separately.

 

Figure 48. Amounts of cations sorbed on cation exchange resins for the
six sampled pedons. Error bars indicate minimum and maximum
values from replicate bags. Note that part A has a different axis scale

 

than other parts. A) Ca; B) Mg; C) Al; D) Fe.

Figure 49. Amounts of cations sorbed on chelating resins for the six
sampled pedons. Error bars indicate minimum and maximum values
from replicate bags. A) A1; B) Fe.

 

Figure 50. The original purchaser of land from the government, T 48 N, R
15 W, Alger County, Michigan. The stump prairie/forest boundary is
shown for ‘

 

Figure 51. The date of the original purchase of land from the government,
T 48 N, R 15 W, Alger County, Michigan. The stump prairie/forest
boundary is shown for reference.

 

Figure 52. Land ownership in 1890, T 48 N, R 15 W, Alger County,

Michigan. The stump prairie/forest boundary is shown for reference. ..........

Figure 53. Per-acre assessed land value in 1890, T 48 N, R 15 W, Alger
County, Michigan. The stump prairie/forest boundary is shown for
reference.

 

Figure 54. Land ownership in 1895, T 48 N, R 15 W, Alger County,

Michigan. The stump prairie/forest boundary is shown for reference. ..........

Figure 55. Per-acre assessed land value in 1895, T 48 N, R 15 W, Alger
County, Michgan. The stump prairie/forest boundary is shown for
reference.

130

131

133

138

140

145

146

148

149

151

152

 

xiii

 

INTRODUCTION AND STATEMENT OF PROBLEM

Logging and uncontrolled wildfires largely destroyed the forests of northern
Michigan in the latter half of the last century (Twining 1983; Benson 1989; Williams
1989). Most of these lands soon regenerated into second-growth forests of successional
hardwoods and pines. In a few scattered locations, however, trees and forest vegetation
have still not returned following the logging and associated fires. One such area, an
outwash plain located in the Upper Peninsula just south of the Lake Superior shoreline, is
known as the Kingston Plains. At one time, this area supported one of the best pine and
mixed pine/hardwood forests of the peninsula (Frederick et a1. 1976), but today the sparse
vegetation consists of reindeer moss, grasses, bracken ferns, and low-growing blueberry
bushes, with a few scattered, stunted trees growing beside the huge pine stumps (“stump
prairie”). The treeless area covers about 2500 ha; it is surrounded by seemingly thriving
second-growth forest in similar positions on similar landforms; some of these are today
being logged a second time.

The objective of this dissertation is to elucidate the factors, both physical and
anthropogenic, that contributed to the existing spatial pattern of regenerated forest vs.
“stump prairie” in the Kingston Plains area. The study focuses on the relationship
between soil properties and vegetation regeneration patterns, but also examines spatial
relationships between vegetation regeneration and 1) pre-existing vegetation patterns; 2)
logging and land management practices; and 3) logging-era fires. A secondary purpose
of the research is to explore the relationship between forest vegetation and soil
development (podzolization) processes in the sand soils of northern Michigan, and
whether, in fact, the long-term elimination of the forest has altered the pathways of soil

development, or even caused soil degradation. Thus the study emphasizes the importance

 

of physical processes and human actions in understanding long-standing environmental

degradation associated with economic development.

Structure of the dissertation

Both the Methods chapter and the Results and Discussion chapter of this
dissertation are structured around three possible factors that could be related to spatial
pattern of forest regeneration in the Kingston Plains: (1) pre-disturbancel forest patterns;
(2) substrate and soil properties; and (3) logging era fires and logging practices. A
section of both chapters is devoted to each of the three factors. Preceding these three
sections, both chapters also contain a section devoted to the current forest/stump prairie

patterns.

 

‘ The term “pm-disturbance” is used in this dissertation to refer to the forest patterns that existed prior to

the logging and associated fires that occurred at the end of the nineteenth century.

 

LITERATURE REVIEW

Stump Prairies

The term “stump prairie” is used in this dissertation to denote a landscape found
on formerly forested land of northern Michigan that now has few trees and abundant old
stumps. Also the term refers to a plant community dominated by grasses, bracken ferns,
and low growing shrubs. Some scattered trees may be present, usually white pine (Pinus
strobus L.), white birch (Betula papyrifera Marshall), black cherry (Prunus serotina
Ehrhart), aspen (Populus spp.), or juneberry (Amelanchier spp.), but a distinctive feature
is the abundance of large, old stumps (now often more than a century old), which attest to
the former forests of these sites. The term itself is used colloquially among land
managers and residents where the stump prairies are common. It has not been formalized
in the scientific literature, although its informal use has been reported (Vogl 1964; Curtis
1971). Rohe (1971 , p. 49) used it to describe an area in northeastern Wisconsin where
repeated fires had lefi a degraded soil supporting only short grasses and some shrubs.
The term has been in use at least since the late nineteenth century, when Roth (1898, p.
13) described “large tracts of bare wastes, ‘stump prairies,’ where the ground is sparsely
covered with weeds and grass, sweet fern, and a few scattering runty [sic] bushes of scrub
oak, aspen, and white birch” remaining on cut-over pine lands in northern Wisconsin.

The vegetation of the northern Michigan stump prairie is similar to that described
in northern Wisconsin as “bracken-grassland” (V ogl 1964; Curtis 1971). Bracken-
grassland communities are dominated by bracken ferns and various grass and forb
species; shrubs such as sweet fern and blueberry are also common (Curtis 1971, pp. 315-

316). Vogl (1964) reported the presence of stumps on all but one of his bracken-

 

grassland study sites. Despite the fact that these sites had been formerly forested and
originated from disturbance, they did not appear to be succeeding back to a forest
community (Curtis 1971, p. 314). A similar community, the blue grass association, was
described as being common on former (pine) forested land in northern Michigan,

although it was being rapidly encroached upon by forest (Gleason 1918).

Possible origins

Little work has been done on the origins of the stump prairies and related
unforested vegetation types of northern Michigan and Wisconsin (Vogl 1964a; Vogl
1970; Curtis 1971). A number of theories regarding their origin(s) have been suggested:

(1) W. Some of these areas lack stumps today and
may not have supported trees, or only scattered trees, before logging (V ogl 1964a; Curtis
1971). Open vegetation2 types similar to that existing on the Kingston Plains were not
unknown in the pre-logging Great Lakes forests, especially on sandy substrates (V ogl
1964b; Vogl 1970; Curtis 1971; Whitney 1986). These “barrens” contained scattered
jack pine clumps and large, open-grown red pine (Pinus resinosa Aiton) trees, and were
maintained by frequent fires (Vogl 1970; Whitney 1986). Many of the areas which
supported only scattered trees in pre-logging times have, however, developed into thick
jack pine forests following the suppression of fires beginning in the early part of this
century (Vogl 1964b; Curtis 1971; Bourdo 1983); in some areas controlled burns have
been used to restore the original savanna vegetation (Vogl 1964b). The “bracken-
grasslands” of northeastern Wisconsin, however, occur today on a variety of pre-logging
site types, ranging from open forests of red pine and scattered white pine to more mesic

stands of sugar maple and associated species, or even boreal forest (Vogl 1964a).

 

2 The term “open vegetation” is used in this dissertation to refer to vegetation communities in which the

forest canopy is not closed, i.e., savanna or grassland vegetation types.

 

(2) WWW- Fire is often cited as a principal cause
of open vegetation types and recurrent fire is thought to be necessary for their
maintenance (V ogl 1964a, b; Curtis 1971). Stump prairie sites may have undergone
particularly intense fires, consuming all or most of the soil organic matter (Curtis 1971;
Frederick et a1. 1976), or particularly frequent fires, killing tree seedlings and eventually
eliminating seed sources (Curtis 1971; Bourdo 1983). Vogl (1970) reported that fire,
while essential to the establishment of northern Wisconsin pine barrens, is less important
than previously thought. He believed that the critical factor in determining their location
was the presence of sandy plains with low fertility that promote droughts and fires of the
proper intensities. Curtis (1971, p. 317) suggested that some of the bracken-grasslands
may have originated following logging and fire, but that fire is not necessary for their
continued maintenance.

(3) WW. Antibiotic production by grassland flora and
competition between grasses, bracken ferns, and tree seedlings might contribute to the
maintenance of open areas once they become established (V ogl 1964a; Curtis 1971).
Reindeer moss lichens are also suspected of allelopathic effects on tree seedlings (Brown
and Mikola 1974; Fisher 1979; Bruhn et a1. 1987). Competition and allelopathy,
however, are unlikely to serve as an explanation for the initial appearance of open
vegetation areas.

(4) Miemflimate. Frequent or prolonged frost activity, or “frost pockets,” might
also inhibit tree reproduction in low-lying or depressional sites, which are common on the
pitted outwash plains where barrens vegetation exists today (Vogl 1964a; Curtis 1971).
Microclirnate, however, cannot explain the occurrence of open vegetation areas on the
relatively flat upland expanses of the majority of the Kingston Plains and other sites
(V ogl 1964a), nor the existence of trees in some nearby depressions.

(5) W. Many authors have suggested that the logging and fires

inflicted on these sites have 1efi the soil too degraded to support contemporary forest

vegetation (e.g., Roth 1908; Rohe 1971; Curtis 1971). Vogl (1964a) thought that many
bracken-grassland sites might have developed “hardpans” while under forest vegetation
which, after elimination of all vegetation in logging and fires, resulted in a harsh
environment fluctuating seasonally from wet to dry, favoring sedges over tree species.
Veatch et a1. (1929), in their soil map of Alger County, distinguished between the soil of
the deforested areas and that where forest exists today, indicating that they believed that

the soil reflected, and may have caused, the vegetation differences.

Role of human disturbance vs. environmental factors
Since before the turn of the century, researchers concerned about the degradation
of Michigan’s forest ecosystems (that is, the change from forest to non- -forested shrub-

and grass-dominated ecosystems) have tended to place primary emphasrs on the role of

 

the logging companies that cut the trees and policies that failed to protect them fiorn, or
even encouraged, widespread fires (Beal 1888; Roth 1906, Buttrick 1921, Curtis 1971).
Certainly 1118;388:1111 activities, as major disturbances to the ecosystem, precipitated the
ensuing forest degradation. Ecosystems are not alike, however, in their ability to
Wimwbances; the concepts of ecosystem resilience and stability have been
formulated to explain the behavior of ecological systems in response to disturbance
(Holling 1973; Barrow 1991). In “natur ” systems, both disturbance and site factors
work together to produce the resulting vegetation patterns (Whitney 1986). For example,
the role of “natural” disturbances in ecosystem maintenance has been much investigated
(e.g., Maissurow 1941; Vogl 1970; Heinselman 1973; Swain 1978), as has the influence
of substrate on forest vegetation patterns (e. g., Livingston 1905; Veatch 1928; Wilde
1933; Brubaker 1975; Whitney 1986; Barrett et a1. 1995). When environmental
degradation is associated with anthropogenic disturbances, however, the tendency has 1)

been to ignore nature’s role in the human/envrronmental interaction equation and place all

._ amt-1‘“

blame on the human side of the equation (Quinn 1991).

This study is conducted on the premise that in order to understand environmental
degradation precipitated by human activity, an examination of the interaction between the
site, the ecosystem, and the activity itself is required. “Environmental susceptibility” has
been defined as a condition whereby the environment of a particular location, because of
that location’s natural features, is especially vulnerable to injury from specific human
activity (Quinn 1991 ). Environmental susceptibility in this sense has been demonstrated
in the severe environmental degradation associated with the copper mining and smelting
activity of the late 19th century in Tennesee’s Copper Basin (Quinn 1991).

Human activity, including logging, removal of the trees, and the subsequent slash
fires, undoubtedly played a role in the change from magnificent pine forest (Frederick et
a1. 1976) to stump prairie on the Kingston Plains. Nevertheless, similar logging and fires
occurred over most of northern Michigan, including large areas of impoverished sandy
soils, and most of these areas today support thriving successional forests of aspen, birch,
pine, and maple. Clearly, some characteristic of the Kingston Plains site itself (and others
like it), something in the manner in which the disturbance occurred and/or the interaction
between the site and the disturbance, is responsible for the stark lack of forest so evident
today. This study examines both site and disturbance factors in order to determine why
this particular place at this particular time experienced such dramatic environmental
degradation. Knowledge about the Kingston Plains site can then be applied to other
regions in order to help managers and planners predict more accurately which activities
might lead to drastic, long-lasting consequences, and which will have relatively minor
effects. Such information is becoming increasingly important as population growth puts
escalating pressure on our natural resources and forest reserves. It can be used to point
out priority areas where disturbance of any kind is likely to lead to severe environmental

degradation from which recovery cannot be expected for decades.

 

 

Pine forests, fire, and succession in northern Michigan

White pine, and to a lesser extent red pine, was an important component of
northern Michigan’s presettlement3 forests. Because white pine is only moderately
tolerant of shade and usually requires exposed mineral soil for successful establishment
(Graham 1941), its widespread presence in the presettlement forest has been interesting to
ecologists. A variety of factors has been invoked to explain the origin and maintenance
of pine forests in northern Michigan. Clearly, either the unique nature of the xeric, sandy
soils or widespread repeated disturbances (fires) arrested the succession of the pine
forests to the more tolerant northern hardwoods types (Whitney 1986).

Early studies often emphasized the role of soils and substrate in determining
presettlement forest type (e.g., Livingston 1905; Veatch 1928; Wilde 1933; Brubaker
1975). The relationship between such physical site characteristics as soil texture,
drainage, and landform type, and presettlement forest composition has been well
established in studies using the notes of the General Land Office surveyors to investigate
presettlement forest composition (Whitney 1986; Leitner et a1. 1991; Barrett et a1. 1995).
It is also possible that more subtle site characteristics, such as sand texture, B horizon
development, or the frequency of finer-textured subsoil bands, may also influence forest
composition (Palik and Pregitzer 1992).

More recently, many have emphasized the role that disturbances, especially fire,
have played in the development of the white pine forests (e.g., Maissurow 1941;
Heinselman 1973; Little 1974; Clark 1989; Frelich and Lorimer 1991). Periodic fires
eliminate the competing shade-tolerant hardwood species and provide a mineral seedbed
conducive to pine seedling establishment. Fire frequency, as determined from the General

Land Office surveyors’ notes, has been related to presettlement forest composition in

 

3 The term “presettlement” is used in this dissertation to refer to the period before major European

settlement of Michigan; i.e., prior to around 1850 in the Upper Peninsula.

 

Michigan. On dry sites in Michigan, presettlement pine forests experienced light surface
fires at intervals of 20 - 40 years, followed by higher intensity fires which killed the
mature pines perhaps every 100 - 200 years (Heinselman 1981). Areas in northern Lower
Michigan experiencing the most frequent fires were forested with jack pine (Pinus
banksiana Lambert) or mixed pine forests, while sites experiencing infrequent fires were
under hardwood forests (Whitney 1986; Mokma and Vance 1989).

Even before the time of European settlement, anthropogenic fires may have
influenced forest vegetation patterns in the Lake States. In northeastern Wisconsin,
savanna vegetation in presettlement times appears to have been closely associated with
Indian sites, and likely resulted from Indian-caused fires (Dorney and Dorney 1989).
Pine forest maintenance in parts of Michigan’s Upper Peninsula was also probably due to

. Indian-caused fires (Loope 1991). In the northeastern United States, however, little
evidence for purposeful burning by Indians has been found (Russell 1983).

Other explanations for the origin and maintenance of the pine forests have also
been advanced. For example, the period during which most of the presettlement pine
forests were being regenerated was warmer and dryer than that prevailing in the mid-19th
century, and may also have had a different fire regime (Clark 1989). Thus, climate may
have played a part in establishing a forest that remained on as a relict as the climate
became cooler and drier. Historical circumstances also influence forest composition and
successional pathways (Palik and Pregitzer 1992).

Most likely, the origin of Michigan’s pine forest can be ascribed to the interaction
of a combination of factors, including fire frequency, climate, soils, human actions, and
historical accident. The dry, sandy soils favor the initial establishment of pine trees,
which in turn contribute to the frequent return of fires, further perpetuating the presence
of the fire-resistant pines (Whitney 1986). Climate and historical circumstances,

' including human actions, further influence the frequency of fire return and thus may also

play a role in determining forest composition (Palik and Pregitzer 1992). A sudden,

 

10

drastic change in certain of these factors, such as happened at the time of nineteenth
century logging, may upset the balance of the interactions, leading to transformations of
forest composition that are dramatic and lasting by creating a new set of equilibrium

conditions (Whitney 1987).

Fire regimes, forest types, and soil development

On the sandy stubstrates of northern Michigan, presettlement fire frequency may
have influenced the degree of soil development by controlling forest composition, which
in turn determines the amounts of organic acids and chelating substances moving through
the solum, i.e., podzolization processes (Mokma and Vance 1989). Mokma and Vance
(1989) found a relationship between B horizon development and forest type in the
northern Lower Peninsula. Weakly developed B horizons are found under jack pine
forests which burned frequently. Recurring fire in jack pine forests (Brubaker 1975;
Sirnard and Blank 1982; Cayford and McRae 1983) results in the burning of the litter
layer and the leaching of fewer organic compounds through the soil, which in turn leads
to slowed development of the spodic (Bs) horizon. Moderately developed B horizons are
associated with red and white pine forests, which burn less frequently than jack pine
forests (Van Wagner 1970; Bourdo 1983; Whitney 1986). The lower burn frequency
creates more available organic compounds, leading to soils with darker Bs horizons and
more iron, aluminum, and organic carbon accumulation. Under northern hardwoods
vegetation, which burns infrequently (Whitney 1986), the strongest soil development is
found, with dark Bhs horizons and the highest organic carbon, iron, and aluminum
accumulations. In this manner soils on surfaces of the same age and texture may develop
to contrasting degrees due to the establishment of contrasting vegetational communities.
It may also be that once the differences in soil development are established they are 1.)

perpetuated because the white pine and northern hardwood species compete better on the

 

11

more strongly developed Spodosols, while jack pine and red pine compete better on the
weakly developed soils (Mokma and Vance 1989).

Many of the relationships between forest species and soil development are
maintained today in the successional forests that followed logging (Mokma and Vance
1989). In some parts of Pictured Rocks National Lakeshore, however, former red and
white pine stands are now dominated by jack pine (Loope 1991). It is not clear whether a
change in the forest type due to logging will, over time, influence soil development by
changing the litter type and fire susceptibility of the site, or whether a change in forest
type alone is sufficient to cause a degradation of an already strongly developed spodic

horizon.

Active soil processes and soil degradation

The close relationship between forest vegetation and podzolization processes in
northern Michigan suggests the question of whether soils, especially spodic horizons,
degrade and become weaker following the destruction of forest vegetation. The spodic
horizon is extremely important to forest yields on the sandy soils of northern Michigan
because it increases water-holding capacity and nutrient availability. The presence and
strength of a spodic horizon can be used to predict site yields for many forest types
(Shetron 1972).

Recent advances in pedogenic theory (Johnson and Watson-Stegner 1987; Phillips
1993a;b) suggest that soils develop along distinct pathways and that these pathways can
be drastically altered by intrinsically-acquired or extrinsically imposed thresholds (Muhs
1984). In the case of the Kingston Plains, the extrinsic threshold has clearly been
deforestation and fire. I have investigated the possibility that deforestation has noticeably
changed the pathways of these soils from one of podzolization (c.f. Schaetzl and Isard i,
1991) to one dominated by low nutrient cycling, organic matter oxidation, and "

predominantly bicarbonate ion weathering (as opposed to organic/fulvic acid weathering).

 

12

Soil development processes and weathering are controlled primarily by the
amount and type of proton donors or acids present in the soil solution (Ugolini and
Sletten 1991). Podzolization is characterized by soil solutions dominated by soluble
organic acids in the O, E, and Bhs horizons, while in the lower solum carbonic acid
becomes the major proton donor at slightly higher pH values (Ugolini et al. 1977, 1988,
1991). By studying soil solution composition, it is possible to determine whether a soil
horizon is currently undergoing the development processes suggested by its morphology,
or whether it reflects former conditions no longer present at the site (Ugolini et a1. 1988;
Ranger et a1. 1991).

The type of proton donor present in the soil solution is thought to be determined
in large part by the vegetation of the site (Ugolini and Sletten 1991). A change of forest
type from deciduous to spruce-dominated has been shown to have affected both the soil
solution and soil properties after a period of only 60 yrs (Ranger and Nys 1994). In
Britain, a change from heather to grasses and bracken has been associated with a
“depodzolization” of the underlying soil (Miles 1985), and also, possibly, a disintegrating
iron pan (Mitchell 1973).

No studies investigating the effect of forest composition changes on podzolization
processes in the northern Great Lakes region have been published, but a few papers have
reported that forest composition does affect spodic horizon morphology. Hole (1975)
observed trends suggesting that spodic horizons in northern Wisconsin gradually fade
upon the removal of hemlock stands. He estimated the “half-life” of a spodic horizon to
be about 100 years after the removal of the trees. Similarly, Milfred et a1. (1967; cited in
Buol et al. 1980) noted much lower B horizon organic matter content in bracken
grassland soils with charred tree stumps than in the adjacent unharvested hemlock stands.
The possibility that spodic horizon expression may have changed since logging is ., ”
especially crucial to this study because at issue is whether soil patterns influenced tree

regeneration or whether tree regeneration patterns influenced soil development. Finally,

13

since soil development, especially in sandy soils, does affect the potential yield of a site
(Shetron 1972; Host et a1. 1987), if the vegetation changes have caused a change in soil
development pathways, they may, by extension, have had “permanent” detrimental

effects on the ecosystem, and, ironically, the soil itself.
Nineteenth century logging in the Upper Peninsula

Logging practices

During the nineteenth century, white pine was the primary target of the US.
lumber industry (Whitney 1987; Williams 1989, p. 198). For this reason, the white pine
forests of Michigan were highly attractive to the lumbering industry. The center of the
pine industry moved from the eastern pine forests of New England into Michigan and the
Lake States beginning in the mid-nineteenth century (Williams 1989, p. 193). Within
Michigan, navigable rivers, combined with the extensive pine forests of the northern
Lower Peninsula, comprised the basis of a flourishing pine industry for about 40 years,
from 1855 to 1895, and about 10 years longer in the Upper Peninsula (Sparhawk and
Brush 1929).

In the early lumbering years, the pine resources of Michigan were believed to be
so extensive that they would never be exhausted (Sawyer 1919). As early as 1885,
however, depletion of accessible pine in the Lower Peninsula began to be noticed
(Sparhawk and Brush 1929). In response, the efforts of the lumber indusuy began to turn
towards the more inaccessible portions of the region, including the Upper Peninsula and
parts of northern Wisconsin and Minnesota (Rector 1953, p. 59).

Advances in lumbering and transportation technology aided in the exploitation of
these areas. As pine stands located near enough to navigable rivers for efficient river
transportation of the logs became increasingly scarce, railroad logging gained in

importance. Construction of logging railroads became common during the 18805, so that

 

14

by 1887 there were 89 logging railroads in Michigan (Williams 1989, p. 212). Railroad
transportation of logs was especially vital in Upper Peninsula lumbering, because large
areas of the forests there were not penetrated by navigable rivers. These areas of forest
were generally the last to be logged (Rector 1953, p. 49).

Increasing mechanization was also introduced in the logging process itself (Rector
1953, p. 59). As accessible lumber became scarce, owners of timber began to consolidate
their holdings and cooperate, with the objective of clear-cutting all the pine and then
abandoning the area. Sometimes tracks would be picked up and re-laid through the forest
as often as yearly in order to quickly clear out the last substantial tree (Williams 1989, p.
230). Regions logged in this manner were efficiently stripped of all pine lumber within a
short period of time, leaving few or no large, standing pine trees.

Pine logging in the Lake States was essentially finished due to lack of pine by
1900 (Williams 1989, p. 228). Only after that time did the lumbering of hemlock and
hardwood species begin to gain in importance (Sparhawk and Brush 1929). As the pine
forests of the Lake States became depleted, many lumbering companies turned their
attention to other parts of the country, especially the Pacific Northwest (Williams 1989,
p. 229).

Logging-related fires

In nineteenth century Michigan, fire followed logging, often as multiple fires in
quick succession. The loggers removed the parts of the tree useful for lumber, leaving
the tops and branches behind as slash. Large amounts of dry slash provided ample fuel
for any fire that was started. Lightning started some fires, but often the most destructive
forest fires were the result of brush burning to clear the land, sparks from locomotives, or
vandals, or berry pickers (Kittredge and Chittenden 1929; Whitney 1987). In a dry year,
smoke from the constantly burning fires in the Lake States could become so thick that it

even became a hazard to navigation on Lake Michigan (Fries 1951, p. 246).

15

By the turn of the century, the destructive nature of repeated fires in the cut-over
lands had begun to be recognized. Sparhawk and Brush (1929) reported that “practically
every acre of northern Michigan [had] been burned over repeatedly during the last 60
years,” and that more than half of the sand plains had been burned at least four times
since they had been cut over. A movement to educate the public about the the destruction
that fire caused and to begin to protect against forest fires began around this time (e.g.,
Sherrard 1903).

Around the turn of the century, some authorities began to associate fire with the
regenerative failure of former pine lands (Whitney 1987). They noted that vast areas of
northern Lower Michigan that had previously been covered by pine trees had no trees left
at all (Livingston 1905). Repeated fires were reported to reduce the cover on these lands
to a few plants which resist burning, such as bracken, bluebem'es, and grasses because the
trees were not able to survive the fires (Sherrard 1903). Filibert Roth (1898, p. 13)
blamed the frequent fires in the pine areas of northern Wisconsin for their transformation
to “wasteland.” The fire was thought to destroy the O horizon of the sandy soils, and
then recur due to the sparse, weedy vegetation growing in the immediate aftermath,
which further reduced the 5011’s fertility (Roth 1898, p. 48). He also pointed out (p. 51)
that it was the lack of burning which allowed the hardwood forests on the more mesic
sites to regenerate quickly after pine had been removed from them.

It is likely that the fire that followed the logging, and not the logging itself, caused
the greater damage to Michigan’s former pine forests (Whitney 1987). The combination
of logging followed by fire altered the disturbance regime and changed the competetive
balance between the forest species. Even though white and red pine depend to an extent
on natural disturbance for their maintenance, the change to a much more intense and
frequent disturbance regime quickly killed both the young seedlings and those mature , ’3’
trees lefi for seed, and favored the reproduction of other species, including aspen and oak

(Whitney 1987).

 

16

Fire following logging has been shown to alter the species composition of forests
in northern Michigan (Whitney 1987). In a stump prairie, however, the forest has simply
failed to regenerate. The origins and maintenance of a stump prairie landscape have been
little discussed in the literature. Soil and substrate patterns, pre-existing forest patterns,
and destruction caused by logging and fire have been suggested as possible factors in
stump prairie formation. In this dissertation, I will investigate the relationship between

these three factors and the patterns of forest regeneration in a stump prairie region.

l.-

'73

K

‘r

.1 1

 

STUDY AREA

General

The study area is located in the northeastern portion of Alger County, Michigan,
bordering on Lake Superior, and centered around the main stump prairie region (Figure
l). The eastern portion of Pictured Rocks National Lakeshore (PRN L) and its buffer
zone adjoining the Lake Superior shore immediately north of the study area. The study
area itself lies mostly within Grand Sable State Forest, with only a small portion under
private ownership. Grand Marais, the nearest settlement, is approximately 20 km to the
northeast. A larger study area consisting of the surrounding six survey townships in
Alger and Schoolcraft counties has been used for those parts of the dissertation in which a
broader geographical perspective was desirable, e.g., the section dealing with
presettlement forest vegetation patterns (p. 54; Figure 2).

The name “Kingston Plains” has long been associated with the stump prairie
portion of the study area, although its use presents some problems because it has
sometimes been applied much more broadly to the extensive outwash deposits extending
from the study area east into Luce county (e.g., Bergquist 1936; Futyma 1981). In this
dissertation I use the term in its narrowest sense, to indicate only the stump prairie

regions located to the southeast of Kingston Lake in T 48 N, R 15 W.

Geomorphology
The topography of the study area is generally level with numerous incised
depressions, in places forming chains of kettle lakes. Prominent escarpments trend east
to west just north of the Kingston Plains stump prairie, separating it from Kingston Lake

to the north. The most recent investigation of the geomorphology of the study area

17

q r . \\\\\\
\\ saflv R .Q .
V s \ \

 

l"

 

 

 

 

 

 

 

 

Figure 2. Map of the six township study area, Alger and Schoolcrafi Counties, Michigan.
Inset square denotes area of field study area shown in Figure 1. Black areas
represent water bodies. Scale l:154,000.

 

20

suggests that the Kingston Plains comprise a large outwash apron (Blewett 1994). The

escarpments result from incision by a series of ice-marginal streams as the ice was

retreating (Blewett 1994); the depressions represent a kettle chain which may be

associated with buried bedrock topography (Blewett and Rieck 1987).

Some controversy exists with regard to the feature mapped as the Munising
moraine, which passes near the study area to the north of Kingston Lake through sections
29 and 30 of T 49 N, R 15 W (e.g., Bergquist 1936). Early interpretations of this as an

ice-contact moraine (Bergquist 1936) were called into question by Drexler et a1. (1983),

'- who suggested that it in fact represents a bedrock high thinly mantled by a pitted outwash
plain. Evidence for the the presence of shallow bedrock in the area is lacking, however,
and drift thickness there likely exceeds 20 m (Blewett and Rieck 1987). Blewett (1994)
interpreted the feature as marking a stagnant ice margin and containing ice-contact and
proglacial stratified drifi. Ice marginal retreat in the study area most probably occurred
within a 300 - 500 year period beginning around 9800 BP (Blewett 1994).

Soils

Well drained, sandy, upland soils in northern Michigan typically develop under an
acid, podzolization regime and are classified as either Spodic Udipsamments or within
Haplorthod Great Groups (Schaetzl and Isard 1991; Barrett and Schaetzl 1992). An old
soil survey of the study area (V catch et a1. 1929) mapped soils of the Rubicon and Au
Train series (today these would be classified as Haplorthods), of sand or loamy sand
textures, having brown or yellow, slightly cemented Bs horizons. A more recent soil
survey of Pictured Rocks National Lakeshore mapped the soils just to the north of the
study area as Rubicon, Kalkaska, and Deerton soils (Entic and Typic Haplorthods; Carey
1993).

21

Vegetation and lumbering

Pre-logging vegetation in the study area consisted of dense forests of white pine
and northern hardwoods, with many large trees 300 - 400 years old (Frederick et al.
1976). As such, this forest was representative of that found on the better developed sandy
soils of northern Michigan (Graham 1941; Bourdo 1983). Although white pine was very
common in these forests, it is not self-perpetuating in upland positions (Graham 1941;
Brown and Curtis 1952). More detail about the pre-logging vegetation of the area will be
provided in the section beginning on p. 54.

The first recorded lumbering operation near the Kingston Plains occurred at
Sullivan’s Landing on the shore of Lake Superior in 1882, and operated for three years,
extracting 50 million board feet of pine (Carter 1967). These logs were stacked on the
beach and rafted to Ontario or the town of Alpena on Lake Huron for milling (Carter
1967). The Kingston Plains area itself was probably logged between 1885 and 1890
(Frederick et a1. 1976), although dates of railway operations suggest that it may have
occurred primarily between 1890 and 1895 (Michigan Railroad Commission 1919). The
Manistique Railway, which would have been used to remove the lumber from the area,
opened in 1886, and was extended to Grand Marais from Seney in 1893 (Michigan
Railroad Commission 1919). Most extensions along this railroad were made between
1889 and 1893 (Michigan Railroad Commission 1919), and abandonments took place in
1895 to 1899, with some in 1903 and 1906. After the largest lumber mill in Grand
Marais closed in 1909 (Carter 1967), the entire line was abandoned in 1910 (Michigan
Railroad Commission 1919).

A crown fire is reported to have come through the area as logging was in progress,
following which logging was accelerated to salvage the burned timber. This fire, in turn,
was followed by several slash fires. The last major fire occurred around 1910 (Frederick

et a1. 1976).

METHODS

Current forest patterns

Stand composition

Seventeen stands of apparently homogeneous vegetational composition and post-
disturbance history as determined from historical aerial photography were chosen for
analysis of tree species composition (Figure 3). Most stands were located near each other
and paired, i.e., a stump prairie and a forested stand. Stands were numbered in the order
of sampling; stump prairie stands were designated with “SAV” (for “savanna”) and
forested stands with “FOR”. Site FOR-7 was included as an alternative pair to site SAV-
7 for the purposes of resin bag placement (discussed below). Areas with atypical
topography or poorly drained soils were avoided, as were regions where trees (usually red
or jack pine) had been planted, whether or not the planted trees had survived.

Stand composition was measured using the point-centered quarter method for
trees with a diameter at breast height (DBH) > 5.0 cm (Cottam and Curtis 1956).
Adequacy of sample size was determined using a species-area curve (Phillips 1959;
Bonham 1989). When > 5 m, distances from the point to the tree were measured using an
optical range finder. Otherwise, a tape was used. Overstory tree species were
characterized for mean diameter, basal area, relative frequency, relative density, relative
dominance, and importance value (Cottam and Curtis 1956; Grieg-Smith 1983). In
stump prairie stands, no tree was recorded for quadrants in which distance to the nearest
tree was > 30 m, the limit of the optical range finder. For such quadrants, 30 m was _ . ’1
substituted for calculations requiring a distance measurement, and “missing” was used 7

instead of a species name in relative frequency and relative density calculations.

22

as. .. «sax
w \ \..\\.

 

24

In many parts of the study area a secondary means of reconstructing the pre-
disturbance forest exists due to the excellent preservation of logging era stumps. For this
reconstruction, it must be assumed that all stumps still remain, and that none have rotted
and disappeared. Many stumps remain in good condition, but some have decomposed
and fallen apart. Wood samples from one well-preserved stump and four decomposing
stumps from various parts of the stump prairie were sent for species identification at the
Forest Products Laboratory (Madison, WI) to assess whether stump preservation may be
related to tree species. The position and size of stumps in all stands for which stumps
were still evident was used to calculate the assumed density and basal areas of trees in the
pre-disturbance stands, using the point-quarter method (Cottam and Curtis 1956).
Distance from the sample point was measured using a range finder (for distance > 5 m).
Otherwise, a tape was used. In areas that are currently forested, locating the stumps in the
field became difficult if no stumps were encountered within a few in of the sample point.
Measurement of distances greater than about 20 m in the forest was also difficult due to
obstructions in the line of sight. Therefore, occasionally a stump could not be located in
a quadrant. In calculating density for these quadrants, 20 m was substituted as the point-
to-stump distance. The diameter at breast height (DBH) for the stumps was estimated to
the nearest 10 cm based on the diameter of the top of the stump (if it was < 1.4 m high)
and the height and preservation of the stump. The diameter values obtained must be
viewed as minimum values, since some stumps, especially of smaller trees, may have
decayed and because bark and outer wood layers were gone.

Age determination of selected large trees was made in order to establish whether
the older trees in the stand had germinated before or after nineteenth century logging
operations. Increment cores were taken and ring counts made in the laboratory. Ring
counts of stumps from large trees felled in recent logging operations were also taken in

the field where possible.

Land cover/stand boundaries

A study of change in vegetation patterns over time, with special attention being
given to the stability of the forest/stump prairie boundary, was conducted using historical
aerial photography. Photographs used for this study include 1939 black and white
panchromatic with a nominal scale of 124,000, and 1954 and 1986 black and white
infrared with a nominal scale of 1:15,840. Change over time between sets of photographs
was determined by comparing representative areas of the boundary at two different dates
at the same scale using a Zoom Transfer Scope. Further examination of the photographs

under magnification was also accomplished using a Delft scanning stereoscope.

Factor 1: Pre—disturbance forest patterns

The primary data source used for the determination of pre-disturbance of Alger
County forest patterns is the records of the General Land Office (GLO) survey. In the
course of establishing section and township boundaries, GLO surveyors noted the
location, diameter, and species of two “witness” trees at each section comer and halfway
along each one-mile long section line (quarter section comer). Also recorded were the
location, diameter, and species of trees located directly on section lines (“line trees”).
Usually four line trees were noted for each mile of section line (White 1984, p. 370).
Although surveyor bias in tree selection is a concern for those using these data in
vegetation reconstruction (Bourdo 1956), GLO witness tree data have ofien been
successfully used to reconstruct the presettlement forest conditions (e.g., Elliot 1953;
Potzger et a1. 1956; Kapp 1978; Delcourt and Delcourt 1977; Leitner et a1. 1991).

GLO survey records (notes) for six townships (T47N, R15W; T47N, R16W;
T48N, R15W; T48N, R16W; T49N, R15W; and T49N, R16W) were obtained from the
archives of the state of Michigan. The surveys of exterior township lines for these
townships were conducted in 1840 and 1841, whereas the subdivision of the townships

into sections was accomplished in 1850 and 1851. All surveyor notations were coded in

 

26

a computer spreadsheet and transferred to the GIS package Arc/Info for analysis.
Surveyor notations coded into the spreadsheet included the common species name,
diameter at breast height (DBH, in inches), and direction and distance (in links) from a
section comer for each tree. Also included in the spreadsheet were comments about
landscape features (e.g., “Enter swamp”) and the surveyor line summary for each section
line, in which the surveyor described timber and soil conditions for the entire mile-long
line. Two types of data were then able to be included in the GIS database: (1) point
locations of trees and with the accompanying descriptive information (hereafter called
“witness trees”), and (2) survey line locations with the accompanying surveyor’s line
summary.

At the end of each mile of section line surveyed, the surveyor was instructed to
provide a description of conditions common along that section line. Included in this
description was typically an assessment of the topography (e.g., “gently rolling”), the soil
quality (e.g., “2” rate sandy”), and the tree species common along the section line, listed
in order of decreasing prevalence (White 1984). Sometimes understory species were also
listed, as were other notable features of the land, especially the presence of swamps, rock
outcrops, burns, or windfalls. Timber species found in swamps were often listed
separately from those in upland portions of the line. These notations were entered into
the database in a separate, miscellaneous “comments” field.

Because the surveyors recorded tree species using the common name of the time,
there are some nomenclatural problems in the data set. Species names used by the
surveyors and their current common and scientific names are given in Table 1.
Occasionally, confusion resulted because the surveyors did not all use the common names
the same way. For example, most surveyors consistently distinguished between white
birch and yellow birch, but at least one surveyor often used the term “birch” without
modification. Where “birch” alone was encountered, it was assumed that the surveyor

was referring to yellow birch.

27

Table 1. Common and scientific names associated with surveyors’ notations of tree

 

 

 

 

species.
Surveyors’ notation Common namel Scientific name1
Alder Alder Alnus spp.
Aspen Aspen Populus spp.
B Ash Black ash Fraxinus nigra Marsh.
B Oak Black oak Quercus velutina Lamarck
Beech American beech F agus grandifolia Ehrhart
Cedar Northern white cedar T huja occidentalis L.
Cherry, B Cherry Black cherry Prunus serotina Ehrh.
Elm American elm Ulmus americana L.
Fir Balsam fir Abies balsamea (L.) Mill.
Hemlock Eastern hemlock T suga canadensis (L.) Carr.
Maple Red maple Acer rubrum L.
S Pine Jack pine Pinus banksiana Lamb.
Spruce Spruce Picea spp.
Sugar Sugar maple Acer saccharum Marsh.
Tamarack Tamarack Larix laricina (Du Roi) K. Koch
W Birch White birch Betula papyrifera Marsh.
W Pine Eastern white pine Pinus strobus L.
Y Birch, Birch Yellow birch Betula alleghaniensis Britton.
Y Pine Red pine Pinus resinosa Ait.
‘ Best estimate.

 

28

Current landcover data (for comparison with the pre-disturbance data) were
derived from the Michigan Department of Natural Resources’ Landuse/Landcover maps
of Alger and Schoolcraft counties (Michigan Department of Natural Resources, Lansing,
MI). The 124,000 maps were made from 1978 aerial photography and were obtained in
digital format. Landuse/landcover classes were simplified by grouping into four broad
categories: (1) upland forest; (2) stump prairie and upland pines; (3) swamp and lowland
forest; and (4) other (Figure 4). Upland areas with pine forest were grouped with stump
prairie because much former stump prairie land had been planted to red pine or other pine
species. The “other” category included beaches, dunes, water features, riverbanks, and
non-vegetated or cultural features.

A line-in-polygon overlay of the surveyor line summaries map and the simplified
landcover map was then performed. The mile-long section lines oflen spanned more than
one landcover category. Therefore, to avoid using the same surveyor comments multiple
times, I analyzed the data using only those lines which had a length greater than 2700 feet
(830 m; or > 1/2 mile) within one polygon afier the overlay. I also eliminated fi'om the
analysis those lines for which the surveyor listed fewer than three tree species in the
summary, since these cases were essentially those for which the surveyor had failed to
provide a line summary (Figure 5).

A point-in-polygon overlay of the witness tree coverage and the simplified
landcover map was also performed. Witness tree data were used to calculate frequency,
basal area, relative density, and relative dominance of the presettlement forest as grouped
by current landcover type (Cottam and Curtis 1956; Delcourt and Delcourt 1974; AuClair
1976; F ralish et a1. 1991). A modified importance value was calculated as the average of
relative density and relative dominance. Relative frequency was omitted from the usual
calculation of importance value because it could not be calculated due to the sampling
design used in the GLO notes: only two trees recorded per corner, with single trees at

intermediate points. Data were analyzed in the units used by the surveyors (inches and

29

t
S
e
r
o

F

 

 

            

                  

 

 

 

.\I

 

                     

   

.‘

\\\'

  

  

. \ ... ..\ \.\\\\.\
$.\ _, \.\\ __\\_.\... m.
s_\.\\ .. _. .m_\=.

. _ m

_ I
__

-\

    

s.\\
I.)
‘2‘\

       

 

 

 

 

Stumps and Upland Pine

Q

 

Figure 4. Map of simplified current landcover classes for six townships in Alger County,

Michigan as of 1981. Inset square denotes area of field study area.

3O

 

           

 

 

 

         

 

I

 

   

     

    

IIIIIIIIIIHIII.II II IIII II.IIIII II
IIIIIII III II I. IIIII . I. IIIII
IIIIIIIII IIII . IIIIII I IIII
IIIIIIIIII IIII
I JIII I
. I I I
I II II II I
.I II IIII I II I
IIflflIIIIIII IIIII/I/ IIIVI a II I IIIIIIIIII~I
I
III III IIIn I II II a a IIII IIII
”IIIuI III IIInoIIIII II I III.IIIII
III I III II

    

 

     

   
  

 

III ’I III
4 I If ~9/Iflui‘fi
.wIIoI II I/II IuIII
. I 0’ III”! .5.” .
.IIIIII ”III. are. III...” .II.II ”II” : I.III.I
..IIII InaIIIIIHIIIhIIIIII~IIIun InII I.III
IIIII III-In .IIII I I. .0IH.IIVIIIIII
III IIIIII~ ~I IIIIIIIuI/IIInIII I
O
. III IIIII I I I o
I III .. a. «I III . I.. .4. . ”IIIIIuI ”IIIIIIIIVII.
II II. .I.IIIII IIIIIIII I.¢III9I .. "LII III III III I
III I II .I.I ”II /I.III.II. II. II IIIIfiI
, I I . . IF
IIII
o I’ll ” III. I’-
o' II
III 0‘ o I S
III III/v
II
I.. I I I ,
I I.. II
II ’I” O s .0 .
IIIIIIIflIIuI IIHIIIIIIIIII . . II: I. ~
IIIIIIIIIII IIII ¢IIIII II .
I III I II IIIII «Io 'l s
O O
IIII .IIu .. .II
II II. I

 

 

 

                    

 

 

                    
          

 

 

 

7

 

 

 

 

ll}

 

 

I
I
II
IInII III
a I IIIII..I
IIst
555.. III

 

 

 

ion lines with at least three timber species listed by the surveyor

Figure 5. Map of sect

in-polygon

ine-

I'all'le areas.

> 1/2 mile) after the 1'

e
3

8 current (1981) stump p

th a length of > 2700 ft. (830 m

andwi

ing represent

overlay. Hatch

31

links), but results were converted to cm and reported as such to facilitate comparison with
data from the sampled modern forest and stump prairie stands.

Individual species relationships with current landcover classes were examined
using contingency table analysis (Whitney 1986; Barrett et a1. 1995). A 2 x 4
contingency table was constructed for each species-landcover variable combination,
where 4 was the number of classes for the landcover variable. Species that occured in
the database fewer than 50 times were eliminated prior to this analysis to avoid errors due
to small sample size. The signed standardized residuals from the contingency table
analysis indicate the direction and strength of association of the species and the current
landcover class; the G2 statistic (G in Whitney 1986) was used to test the statistical
significance of the association between a species and landcover class (Whitney 1986;

1990; Barrett et a1. 1995).

Factor 2: Substrate and soil properties

Field methods

Within each of the 17 sampled vegetation stands (9 forest and 8 stump prairie), a
representative pedon was chosen for sampling. A pit, approximately 1.5 m x 0.7 m and
deep enough to expose the C horizon, was dug. Standard field descriptions were made
(Soil Survey Staff 1981) and bulk samples (about 0.5 kg) were taken by genetic horizon
for laboratory analysis. The approximate proportion of ortstein in each subhorizon was
estimated by measuring the horizontal extent of cemented material exposed on a 1 m
wide section of the pit face. Ortstein portions of B subhorizons were sampled separately
when strongly cemented ortstein comprised a significant proportion (> 20%) of the
subhorizon. Deep sampling (> 2m) by bucket auger was performed to check for the
existence of fine-textured lamellae (Schaetzl 1992) below the base of the pit, which have
been shown to affect forest growth (Miles and Franzrneier 1981; Host et a1. 1988).

As a further means of measuring soil development near the stump prairie/savanna
boundary, six transects perpendicular to the trend of the boundary were made (Figure 6).
Beginning at the approximate location of the boundary, six 1 m2 quadrats were
established at 30 m intervals on each side of the boundary. Within each quadrat, nine
regularly spaced samples were taken with a push-probe to a depth of approximately 50
cm or until ortstein obstructed the probe. At each sample point, I recorded whether
ortstein was present or absent, and also the Munsell color of the darkest, reddest soil

retained in the probe, usually from the upper B horizon.

. ow \\ . \N\\\\\\\\

. In. \\ .
\R is. \k \\\x\

W \4M\\\\

 

 

.\

If;

V.

91-:

”-11

(r. 9

1
I,“

34

Laboratory methods

Laboratory analysis of the horizon—based samples (including ortstein as a separate
subsample) included pH (2:1 in water and KCl) and particle size distribution by pipette
(Soil Survey Laboratory Staff 1992). Organic carbon (OC) content was determined
using a modified Walkely-Black procedure (Singer and Janitzky 1986). Extractions for
several elements were performed using acid ammonium oxalate (Feo, A10, and Sio),
sodium pyrophosphate (F 6,, and A1,), and sodium citrate-dithionite (Fed and A1,]; Soil
Survey Laboratory Staff 1992). Extracts were analyzed by DCP spectroscopy in the
MSU Department of Crop and Soil Sciences laboratory. Optical density of the oxalate
extract (ODOE) at 430 nm was determined on a Bausch and Lomb Spectronic 20
colorimeter (Daly 1982).

For depth plots and calculations requiring a single value per horizon, B
subhorizon values were calculated by weighting the ortstein and non-ortstein subsamples
according to the field estimate of ortstein proportion of the subhorizon. For example, for
a BS] horizon with an estimated 30% ortstein content and 2.9% clay in the matrix and
2.6% clay in the ortstein portion, the weighted le clay content would be calculated as
[(2.6 x 30) + (2.9 x 70)] / 100, or 2.8%.

To facilitate comparison of pedons with different horizon sequences and horizon
thicknesses, a weighted B horizon value was calculated by using the mean value for all B
subhorizons (not including the BC horizon), weighted for (multiplied by) subhorizon
thickness using the following formula: 2 (P * H) / 2 H, where P = the soil property for a
given subhorizon and H = horizon thickness in cm. Weighted solum values were
calculated analogously to the weighted B horizon values, but including all mineral

subhorizons above the upper C horizon.4

 

4Solum weighted values include BC horizon values because (1) for most pedons, the difference between B

horizon weighted values and solum weighted values would simply be the B horizon in the solum

35

In situ monitoring of soil processes

Three pairs of pedons from the 17 pedons described and sampled above were
chosen for an in situ study of podzolization processes using buried resin bags (Righi et a1.
1990; Ranger et a1. 1991; Ranger and Nys 1994): sites SAV-6 and FOR-5; SAV-7 and
FOR-7; SAV-8 and F OR-95 (Figure 3). Site FOR-7 was substituted as a partner for site
SAV-7 after its original pair, site FOR-8, was discovered to have atypically cemented B
subhorizons and a coarser sand and gravel immediately below the solum. A Na-saturated
cation exchange resin (Amberlite IRN 77; Rohm and Haas Co., Philadelphia) and a Na-
and H- saturated chelating resin (Chelex 100; Bio-Rad Laboratories, Richmond, CA)
were used; comparison of the relative amounts of Fe and Al adsorbed on each resin gives
information on the speciation of Fe and Al in the soil solution for the period of burial.
Small bags (approximately 5 x 8 cm) containing 5 g (moist) of a resin were constructed
of nylon tricot fabric with mesh small enough to retain the resin. One bag of each resin
type was inserted into the soil from the face of the soil pit at the top of the E, in the
uppermost B, and in the BC horizons. Four replicate sets of these bags were installed in
each of the six pedons. Each pit was then refilled. Bags were buried in August and
September, 1994, and retrieved in late May, 1995. Following retrieval, bags were air-
dried, and the resin removed and weighed. Cations were desorbed from the resins (about
1.2 g of dry chelating resin and about 3.0 g of dry cation exchange resin each") by
shaking for 4 h in 40 m1 of 1 N HCl for the cation exchange resin and 1 N HNO3 for the

 

weighted value if the BC horizon were not also included; and (2) the BC horizon, though
transitional to the C horizon, is more like the B horizon than the C horizon in these soils and should

be considered part of the solum. In some pedons, ortstein columns continued into the BC horizon.

SSite numbers indicate the order in which vegetation sampling was completed; paired pedons for soil resin

bag samples do not necessarily share the same number.

6Differences in dry weight of chelating and cation exchange resins are due to differences in the original

moisture content of the resins. The entire amount of resin was used in the analysis.

36

chelating resin. The extracts were analyzed by DCP spectroscopy in the MSU

Department of Crop and Soil Sciences laboratory.

Factor 3: Logging era fires and logging practices

Historical records relating to land ownership, property tax assessment, and tax
payment during the logging era were examined for evidence of whether logging practices
in the stump prairie areas were different from those in the forested areas in either
character or timing. Records were obtained from the archives of the State of Michigan
and included: tax/assessment rolls of Alger County for the years 1885-1887, 1890, 1895,
and 1899; abstracts of sales of state tax land for the years 1888 and 1892-1894; and the
land tract book which records the initial transfer of property in Michigan from public to
private hands. From these sources, data for T 48 N, R 15 W, which approximately
coincides with the field study area, were input into the GIS for mapping. Mapped data
include: the name of the original purchaser of the land; the date of the original
acquisition; the name of the owner for the years 1890 and 1895; and the per-acre assessed

value for the years 1890 and 1895.

RESULTS AND DISCUSSION

Current forest patterns

Stand composition

The forests surrounding the Kingston Plains continue to be used as a source of
timber. Most forests have been logged at least once since the turn of the century, and
many show signs of having been logged more than once. Decomposing hemlock stumps
present in many of the forested stands point to selective extraction of the larger hemlock
trees 30 to 40 years ago. Current logging operations are removing mostly hardwoods
such as red maple, beech, and white birch, as well as the larger white pine trees.

Modern logging has also taken place in parts of the stump prairie itself, as
evidenced by the presence of relatively undecomposed white pine stumps and branches
alongside the more weathered, nineteenth century stumps. Many parts of the stump
prairie have been artificially planted to red pine and jack pine, especially in the western
portion of the study area. (These areas were avoided in sampling.) Observations in and
around the plantations suggest that natural red pine and jack pine recruitment is not
common. White pine seedlings predominate, even in the plantations. Prescribed burning
for management of sharptail grouse has been practiced in the northeastern corner of the
Kingston Plains (parts of section 10, T 48 N, R 15 W). Site SAV-4 is located within this
controlled burn area.

In forested stands today, red maple is the most important species, followed by
white pine and beech (Table 2). Hemlock and balsam fir are also common. The species
composition is quite consistent from one stand to another (Table 3), with the exception of

stand FOR-6, which has an unusually high importance value for white pine, and a low

37

38

Table 2. Diameter at breast height (DBH) and importance values by tree species for
forest and stump prairie areas.

 

 

 

 

 

 

Species N Mean Max Total Total Rel. Rel. Rel. Imp.
DBH DBH DBH basal Dens. Freq. Dom. Val.
area
2111 cm‘ 942
Forest
Red Maple 148 19.1 48.5 2829.4 53916.5 40.2 35.3 39.7 38.4
White Pine 66 17.8 68.2 1173.9 24198.3 18.2 15.9 17.8 17.3
Beech 56 18.7 60.0 1044.5 25660.6 15.2 17.4 18.9 17.2
Hemlock 32 19.1 65.2 612.4 14941.0 8.7 10.1 11.0 9.9
BalsamFir 44 9.2 28.0 404.6 3606.3 12.0 13.0 2.7 9.2
White Birch 7 40.6 60.0 284.4 9887.1 1.9 3.4 7.3 4.2
RedPine 10 19.7 36.7 196.5 3513.1 2.7 3.4 2.6 2.9
PinCherry' 2 6.1 6.1 12.1 57.5 0.5 0.5 0.0 0.4
Juneberry 1 10.2 10.2 10.2 81.7 0.3 0.5 0.1 0.3
Aspensp. 1 8.2 8.2 8.2 52.8 0.3 0.5 0.0 0.3
White spitee' 1 5.6 5.6 5.6 24.6 0.3 0.5 0.0 0.3
Total: 368 17.9 68.2 658181359395
Smmn
E ..
White Pine 151 18.5 103 2793.2 70714.4 34.6 29.9 79.0 47.8
Missingz 170 0 0 0 0 39.0 29.9 0.0 23.0
RedMaple 32 15.9 36.5 509.3 7128.6 7.3 10.7 8.0 8.7
BlackCherry 23 8.0 13.5 183.6 1263.3 5.3 8.4 1.4 5.0
JackPine 14 15.8 20.5 220.7 2968.4 3.2 5.1 3.3 3.9
Juneberry 17 6.3 8.6 107.6 554.3 3.9 6.1 0.6 3.5
BlackSpruce' 11 10.0 19.7 110.5 1021.9 2.5 3.7 1.1 2.5
RedPine 5 29.3 35.1 146.4 3455.6 1.2 1.9 3.9 2.3
White spmoe' 4 21.7 35.0 86.6 1864.1 0.9 1.4 2.1 1.5
Aspen sp. 7 7.1 7.7 50.0 281.2 1.6 2.3 0.3 1.4
Balsam Fir 1 6.2 6.2 6.2 30.2 0.2 0.5 0.0 0.2
BlackOak' 1 18.3 18.3 18.3 263.0 0.2 0.0 0.3 0.2
Total: 436 15.9 103.0 4232.4 89545.1

 

7 Pin cherry: Prunus pensylvanica L.; Black spruce: Picea mariana (Miller) BSP; White
spruce: Picea glauca (Moench) A. Voss; Black oak: Quercus velutina Lamarck.
2 Denotes a quadrant in which no tree was located within 30 m of the sample point.

39

Table 3. Importance values of tree species for the nine sampled forest stands'.

 

All
FOR-1 FOR-2 FOR-3 FOR-4 FOR-5 FOR—6 FOR-7 FOR-8 FOR-9 forest

 

stands
Aspen 0 0 0 0 0 2.3 0 0 0 0.3
Balsam Fir 25.0 4.6 9.1 4.2 17.2 0 8.7 18.8 3.9 9.2
Beech 12.2 33.4 23.6 27.7 5.1 0 14.7 13.8 21.6 17.2
Hemlock 10.1 2.3 16.3 18.5 8.0 O 20.3 0 12.4 9.9
Juneberry 0 0 0 0 O 0 0 0 2.0 0.3
Pin Cherry 0 0 0 0 3.4 0 0 0 0 0.4
Red Maple 25.3 47.8 48.6 28.5 58.7 3.7 37.5 64.8 31.3 38.4
Red Pine 0 0 0 0 0 26.7 0 O 0 2.9
White Birch 0 6.5 0 4.7 0 4.0 16.4 0 0 4.2
White Pine 27.42 5.5 2.3 16.5 7.6 63.2 2.4 2.6 28.8 17.3
White Spruce 0 0 0.1 O 0 0 0 0 O 0

 

lTrees found on the stump prairie but absent at forest sample sites include: black cherry,
black spruce, jack pine, and black oak.
2The tree species with the highest importance value in each stand is highlighted in

boldface type.

40

value for red maple (Table 3). It is also the only forested stand in which red pine is
present. These anomalies may be because F OR-6 occupies the site of a former stump
prairie where the trees have recovered enough to allow the canopy to begin to close, as
discussed below, and therefore retains the species composition of its stump prairie
origins.

In stump prairie stands, the living trees are strongly dominated by white pine
(Table 2). The lack of trees is shown by the fact that the most common point-quarter
observation on the stump prairie (N=170) is the absense of a tree within 30 m of the
sampled point (Table 4). Red maple and black cherry are also common in the stump
prairie stands. Except for these most common stump prairie species, however, the species
composition from one stump prairie stand to another varies markedly (Table 4). For
instance, juneberry has a high importance value (27.9) in stand SAV-2, but is either
absent or has very low values elsewhere. Similarly, black cherry is remarkably important
(22.6) in stand SAV-8 (Table 4). This variability in species composition suggests that the
chance circumstances of seed or sprout availability has been a prominent factor in
determining the recovery and composition of the stump prairie areas. Black cherry seeds
commonly fall to the ground in the immediate vicinity of the parent tree, except for those
few spread by birds or omnivorous mammals (commonly bears on the Kingston Plains) in
their droppings (Marquis 1990). Black cherry also readily resprouts from its stumps
(Marquis 1990). Juneberry trees also have large fruit which may be spread by birds
(Barnes and Wagner 1981)

As expected, absolute density of the trees is much greater (nearly 10x) in the
forested stands than in the stump prairie stands (Table 5). Stand FOR-6 is the single
exception to this. Stands SAV-l and SAV-6 have the highest absolute densities among
the stump prairie stands (Table 5). These two stands are similar to FOR-6 in that the

trees in these regions have become quite large. In all three stands, white pine has very

41

Table 4. Importance values of tree species for the eight sampled stump prairie standsl.

 

 

A11

stump

SAV-l SAV-2 SAV-3 SAV-4 SAV-S SAV-6 SAV-7 SAV-8 Prame

stands
Missing2 0 37.2 36.8 63.8 33.3 9.6 4.6 12.7 23.0
Aspen 3.2 O 0 O 0 0 0 8.1 1.4
Balsam Fir 0 0 0 0 0 0 1.6 0 0.2
Black Cherry 5.0 10.9 5.0 0 0 2.3 1.6 22.6 5.0
Black Spruce 9.9 O 0 0 4.9 0 3.4 O 2.5
Jack Pine 13.3 0 0 0 0 0 12.8 0 3.9
Juneberry 0 27.9 0 0 5.5 0 1.6 1.8 3.5
Red Maple 0 20 0 36.2 20.6 0 14.1 32.2 8.7
Red Pine 8.5 0 5.9 0 0 2.8 0 5.3 2.3
White Pine 60.13 3.9 52.3 0 35.8 85.2 53.1 15.4 47.8
White Spruce 0 0 0 0 0 O 6.3 1.9 1.5
Black Oak 0 0 0 0 0 0 1.0 0 0.2

 

lTrees found commonly in the forest but absent at stump prairie sample sites include:

beech, hemlock, pin cherry, and white birch.

2 Denotes a quadrant in which no tree was located within 30 m of the sample point.
3The tree species with the highest importance value in each stand is highlighted in

boldface type.

42

Table 5. Absolute density of trees in the 17 sampled stands.

 

 

 

 

Stand-ID Absolute Density
119251181

F OR- 1 4258
F OR-2 2299
FOR-3 3247
FOR-4 3045
FOR-5 3 361
FOR-6 34 3
FOR-7 2783
FOR-8 2465
FOR-9 3512

Forest mean: 2816
SAV—l 657
SAV-2 26
SAV-3 25
SAV-4 13
SAV-S 90
SAV-6 1284
SAV-7 73
SAV-8 97

 

Stump prairie mean: 295

 

p v

A“

 

43

high importance values (FOR-6, 63.2; SAV-l, 60.1; SAV-6, 85.2), red pine is present,

and red maple is rare or absent (Tables 3 and 4).

44

Land cover/stand boundaries

Following the logging of the nineteenth century, vegetation patterns in the
Kingston Plains area have not been static. The purpose of this section is to consider the
changes in forest patterns evident in historical aerial photography, with particular
reference to the boundary between stump prairie and forest. This discussion will focus on
data compiled from the 1939 and 1986 photographs, although photographs from 1954
were also included in the analysis. Three questions will be addressed: (1) Has the nature
of the forest/stump prairie boundaries changed over time? (2) Has the boundary moved
over time? and (3) Has the character of the stump prairie itself changed?

W In the study area, the transition between open stump
prairie and closed-canopy forest usually occurs within the space of a few meters, with
little apparent intermediate landcover between them. Truly discrete boundaries, however,
were more widespread in 1939 than in 1986. In 1986, in some places a more gradual
transition from stump prairie through scattered trees to closed-canopy forest existed,
especially along the western, northwestern, and southwestern edges of the study area
stump prairie (Figures 7 and 8). The eastern boundary tended to remain more discrete
(Figure 9), possibly because a controlled burn was conducted in 1983 in the northeastern
corner of the stump prairie.

W The overall pattern of stump prairie and forest in the
study area remained stable over the nearly 50 year time span from 1939 to 1986.

Nevertheless, a close examination of the photography reveals significant7 movements of

 

7 The intent of this section is to provide a qualitatixe description of changes in boundary location over
time. No guantitatixe measurements of boundary movements were made, due to the inaccessibility

of the necessary aerial photography at the time revisions to this dissertation were accomplished.

45

2)

 

Figure 7. Aerial photographs of the study area, sections 29 and 30 of T 48 N, R 15 W. l)
1939; 2) 1986.

46

 

Figure 8. Aerial photographs of the study area, sections 8, 9, l6, and 17 of T 48 N, R 15
W. 1) 1939; 2) 1986.

47

 

)
2

,R15w. 1)

Figure 9. Aerial photographs of the study area, sections 10 and 11 of T 48 N

1939; 2) 1986.

'II

t]!

10

If)

48

the boundary in many locations, usually with the forest gradually encroaching upon the
stump prairie area. For example, in the southwestern comer of the study area the forest
had noticeably moved into former stump prairie regions between 1939 and 1986 (Figure
7). This is especially obvious around the small “peninsula” of stump prairie that extends
south of the main road intersection (Figure 7, location A ). In 1939 this peninsula was
quite wide, but by 1986 the forest had nearly engulfed the road. Similar shifting of the
boundary is apparent throughout the region visible in Figure 7.

Farther to the north in the study area, an even more dramatic advance of the forest
boundary can be seen (Figure 8). In the 1986 photo, the forest had expanded to include a
few evergreen trees that in 1939 were isolated, far from the forest itself (Figure 8,
location A). The 1939 forest boundary is still apparent in the 1986 photo as a change of
texture in the forest. Examined with a stereoscope, the 1986 photos reveal that the trees
in the original 1939 forest area are taller and closer together than in the newly expanded
portion.

Smaller clumps of trees that had been isolated from the main forest in 1939 also
grew in size such that by 1986 they had often become joined, or nearly joined, with the
main forest itself. Figure 7 shows such an isolated forest “island” in 1939 (location B).
In the 1986 photo the area between it and the main forest had nearly filled in. Similarly,
isolated trees on the stump prairie in the 1939 photos often had become the center of
clumps of trees by 1986. For example, in 1939 a small group of individual evergreen
trees (probably pines) was located just north of the road (Figure 7, location C). In the
1986 photo these trees had become the nucleus for a much larger clump of mixed-species,
closed-canopy forest. The formerly scattered trees apparently provided an environment
favorable for the establishment of both evergreens and hardwoods, as well as a source of
seeds and sprouts.

Although evidence for the expansion of the forest boundary between 1939 and

1986 is widespread and no example of boundary retreat could be found, the amount of

49

change varies from location to location. In the northeastern comer of the Kingston Plains
the forest boundary had moved only slightly (Figure 9). In this region the stump prairie
has been managed for sharptail grouse, including a controlled burn in the early 19805.
Even here, however, the forest boundary has measurably advanced into the stump prairie.
Perhaps the most stable boundary in the study area can be found at its southeastern comer
(Figure 10, location A). Very little movement of the main boundary between the two
dates can be detected in this view. The clumps of trees along the old railroad grade in the
north part of the photo had also expanded only slightly. Under the stereoscope, however,
the trees between the lake and the wetland to its northeast (Figure 10, location B) are seen
to be small and fairly open in the 1939 photo. Much thicker foliage and a closed canopy
can be seen in the 1986 photos of the same area, even though the location of boundary
itself has not changed significantly.

Around some of the kettle depressions on the Kingston Plains the stump
prairie/forest boundary has apparently been affected by the topography. Trees usually
establish more slowly within the depressions than the surrounding higher area. Between
1939 and 1986, however, in some places the forest boundary had advanced farther into
such depressions. The depression shown in Figure 11 is deep with relatively steep walls;
an old railroad grade follows the bottom of the depression. On the west side of the
depression (Figure 11, location A), which is the steeper of the two sides, the trees had
advanced slightly into the depression by 1986. On the east side, however, the forest
boundary had advanced much farther (Figure 11, location B). Very few trees can be
found near the bottom of such depressions, even though stumps are common, indicating
that they were once forested.

MW Comparison of the 1939 and 1986
aerial photography also reveals that the character of the vegetation on the stump prairie
has changed. In 1939, the stump prairie in most places was nearly devoid of trees, and

only very scattered evergreen trees can be found at all (Figure 12). Under close

50

 

sections 27 and 28 ofT 48 N, R15 W.

10. Aerial photographs of the study area,

1) 1939; 2) 1986.

Figure

51

 

Figure 11. Aerial photographs of the study area, sections 8 and 17 of T 48 N, R 15 W. 1)

1939; 2) 1986.

52

2)

 

Figure 12. Aerial photographs of the study area, sections 20 and 21 of T 48 N, R 15 W.
1) 1939; 2) 1986.

53

stereoscopic examination, occasional snags are visible in the stump prairie, but usually
the only features are small, dark dots interpreted as stumps. By 1986, however, most
parts of the stump prairie contained at least some scattered trees. Many of the trees
visible in the 1986 photographs are hardwoods, not evergreens. In places where scattered
evergreen trees had been visible in 1939 (Figure 12, location A), the stump prairie had by
1986 become an open-canopy forest. Even where trees had been eliminated by controlled
burning some trees are present in 1986 that had not been there in 1939 (Figure 9, location
A). The density of trees seen in 1986 photos on the stump prairie varies from place to
place. Fewer trees are apparent in the eastern portion of the plains (Figure 10), in
depressions (Figure 11), and in the controlled burn area (Figure 9) than in the western
portions of the study area (Figures 7 and 12).

Examination of the aerial photography in 1939, 1954, and 1986 reveals that the
stump prairie has changed naturally over the 50 year timespan. The forest boundary is
gradually encroaching upon the stump prairie, and the stump prairie itself is gaining more
and more trees. This suggests that the ecosystem is rebounding from the disturbance of
logging, though slowly. Except for where management practices such as controlled burns
or new disturbance prevent it, with the passage of another 50 - 100 years, the Kingston

Plains will probably become a forest once again.

54

Factor 1: Pre-disturbance forest patterns

Surveyor line summaries

The surveyor line summaries were subjective on the part of the surveyor, but they
do provide a secondary means of assessing forest conditions at the time of the survey, and
have been used to support information determined from witness tree distributions
(Bourdo 1956; Frederick et a1. 1976). Surveyor line summaries have been used most
often in determining disturbance frequency in the presettlement forests, based on the
notations of fires and windfalls (e.g., Canham and Loucks 1984; Whitney 1986).

The frequency of citation in line summaries of individual tree species for each
current landcover category indicates a spatial correspondence between current landcover
classes and the surveyors’ subjective assessment of the forest (Table 6). In upland forest
areas, sugar maple was most commonly cited first, with beech and hemlock also ofien
cited first (Table 6, Figure 13). Yellow birch was more often placed in the second or
third position, as were beech and hemlock. Balsam fir and red maple were common
associates, listed often in third or fourth position.

For most lines in current stump prairies, the surveyors overwhelmingly recorded
white pine as the predominant species (Table 6, Figure 14). Hemlock and red pine were
often cited as second in prevalence, while beech and red maple were usually listed third
or fourth (Table 6).

Surveyor line summaries in swamp areas are more difficult to interpret, in part
because the surveyors listed timber on upland parts of a line separately from timber in
parts of the line that were swamp. Swamp species were in these cases included in a
separate field in the database, which was not included in this analysis. Thus some of the

timber listed for these section lines may actually correspond to the timber in adjacent

55

Table 6. Citation frequency of the tree species in surveyors’ line descriptions for three
current landcover classifications.

Position in abundance lists

 

 

 

Smcies - lst ‘ 2nd . 3rd % 4th
1311 Percent 1111 Percent N1 Rascal NF Kareem
Upland forest
SugarMaple 35 28.5 15 12.2 10 8.1 3 2.4
Beech 22 17.9 32 26.0 20 16.3 17 13.8
Hemlock 21 17.1 18 14.6 14 11.4 15 12.2
White Pine 17 13.8 8 6.5 3 2.4 16 13.0

YellowBirch 13 10.6 24 19.5 29 23.6 28 22.8

Red Maple 8 6.5 13 10.6 22 17.9 14 11.4
Balsam Fir 6 4.9 11 8.9 21 17.1 24 19.5
WhiteCedar 1 0.8 2 1.6 1 0.8 2 1.6
Spruce 0 0.0 0 0.0 2 1.6 1 0.8
BlaskClmy 0 0.0 0 0.0 1 0.8 0 0.0
Tamarack O 0.0 0 0.0 0 0.0 2 1.6
Stump Prairie
White Pine 41 77.4 2 3.8 3 5.7 3 5.7
Hemlock 6 11.3 19 35.8 8 15.1 6 11.3
Red Pine 1 1.9 12 22.6 2 3.8 1 1.9
Red Maple 1 1.9 4 7.5 12 22.6 7 13.2
Tamarack 1 1.9 1 1.9 0 0.0 1 1.9
Jack Pine 1 1.9 0 0.0 2 3.8 0 0.0
WhiteCedar 1 1.9 0 0.0 0 0 0 O 0.0
SugarMaple 1 1.9 0 0.0 0 0.0 0 0.0
Beech 0 0.0 5 9.4 13 24.5 9 17.0
YellowBiich 0 0.0 4 7.5 2 3.8 10 18.9
Balsam Fir 0 0.0 3 5.7 6 11.3 7 13.2
Spruce 0 0.0 3 5.7 4 7.5 6 11.3
Alder 0 0.0 0 0.0 l 1.9 l 1.9
AmerimElm 0 0.0 0 0.0 0 0.0 1 1.9
Swamp
Hemlock 10 22.2 3 6.7 5 11.1 3 6.7
White Pine 9 20.0 2 4.4 2 4.4 5 11.1
WhiteCedar 7 15.6 1 2.2 2 4.4 1 2.2
Spruce 5 11.1 10 22.2 5 11.1 3 6.7
Tamarack 4 8.9 7 15.6 5 11.1 0 0.0
Beech 4 8.9 2 4.4 2 4.4 7 15.6
Red Maple 3 6.7 2 4.4 3 6 7 6 13.3
Balsam Fir 2 4.4 1 2.2 6 13.3 4 8.9
SugarMaple 1 2.2 3 6.7 2 4.4 2 4.4
YellowBirch O 0.0 1 1 24.4 6 13.3 1 2.2
Red Pine 0 0.0 3 6.7 0 0.0 1 2.2
WhiteBirdi 0 0.0 0 0.0 1 2.2 1 2.2
Alder 0 0.0 0 0.0 0 0.0 2 4.4
Black Ash 0 0.0 0 0.0 0 0.0 1 2.2

 

rNumber of section lines occurring in landcover classification where species was listed by
surveyor in given position.

i
11"

‘z‘i‘l
~41

56

                                     

 

 

 

 

 

I
IIIIIIIIIIIIIIIIIIII
/IIIIIIII III III

IIIIIIIII III IIIIIIIII III~
IIIIII IIIIIII»II
IIIIIIIIIII IIIIIIII»
.I IIIIII IIIII
..II».IIIIIIII IIIIG

II »III IIIIII IIIIII
II»I IIIII
II II

II.II
aII I66“

— I». »IIIIIII IIII»II»II

Immm I

 

I.II... ”WW
III
~IIIIIIII OI’II
¢IIIII’IIIII *IIIW.” II IIIIII ~* ¢ I.IM6 0:
II»II III»I.IIIIIIm.II

+5.99 IIIIII I$III IIII¢

IIIII .IIIII.IIIII I IIIIIIIIIIII»III I
II.IIIIIIII III III IIQI I QII
.IIIIII9

. OUII6’.’ I ’

II.II»..I.»III.II.».I.»IVIWIII»III

was?»

4 I I.IIIIII»I»

III...I .

I. III I I
1 I IIII IIII I.I
I »I Im .IIIII..I.I IIII III»I. .I.III IIIIIIII»IIII III IIIIIII I. III
III I II II I I I I I II.II.» I.»
III I MII ..I.. II.I II.IIIIII I»IIIIII II IIIIIIIIIII II»
II
II

III II.
IIIIIIMIIIIMIIIIIMI IIIHIIGIIIIIIIII IIII IVEI

“.I.IIIéwII

4.
»»>III.II.»IIIH»I»»IIIIAI..

I...I»I»IIIII»I

I»II.

III: II I IIII. 4 IIIII...

"I! IIIIIIIIII» II III.I»II».VI I.I” II 01.0%»!

III/IIII IIIIfla I‘I

wIIIIIIIII IIIIII»» II»IIII mafia
II»IIII

IIIIIII IIIIIw. IaIIIII

III

I
III I».» I»II
I‘ IM:III”WHHIIIII

 

..I/{I.. . I II III
...» »»IIIIIIIIIIII I»I/III .IIIII..I —I .I».I». I.IIIII»—'
.. ..I.. ...... ..I.”. ....
II I I
III»IIIIIII ImII»I

I... _

1.....1. ...- ._

I’IIIII “
l I f

 

 

 

 

 

 

 

 

 

 

 

 

rairie areas.

(short dashes), or hemlock (long dashes) as the most abundant timber type.

Figure 13. Survey lines for which the surveyor listed sugar maple (solid lines), beech
Hatching represents present stump p

57

 

.... I.. m I
I I I
II I‘flI
”II.Ifl I.IIIIIIMIIIII II
I I. I
II .II »II III» III I
III II I ..
IIIIII I I
.II IIIIIIIIII II»II II»II ..I. 2m
.I.» I.IIIIm ... ..I .. .I....II
I .I II II.- "II IIIIIIIII “fl” IIIII I:¢I .I
..I. .II »I » I III.»
IIII III» I I
o. I. I II .I “I.
III III I IIIII II cow-.II I
. . I
III III II
IIII III . 9II
..II
.I.IIIWI II 6: II.I»II IIIII .II »III I . III “III ...
I. .III.IIIII II I III I IIIIIII I
.III.II.»II IIIIIIIIII . III” I I III ”II.I/III III II.II.
. »I.I III »I III.I IIII IIIII II I.I II
I II I II I . I I .
IIIIII I II. II II
IIIIIIIIIIIIIIIIIIIIIIIIIII ». IIII. ..II .»
IIIIIIIIIII IIIIIIIIII IIII III.. .. I.
IIIIIIIIIIIII IIIIIIII III . II .
.IIIIIIIIII IIIIIIII I. I9 .I
II I
o I IIIII I I I I‘I‘
Ida/“III”. IIIIMIII II IIIIIIIIIII IIII IIII .IIva
III»» IIIIIII IIII I III/II.I w I I .IIII» IIIIIhI III .I. ..I. ..
II I I. I I
......I»... .............. .. 2...... ...... .
II I IIIII II » I I
I»IIIII I I IIIIII I
’I I III III III II%
III »» III I.III»»III
IIIIII IIIIIIIIIIIIIII .I
III ”II |
II
I
I ”HMIII I.
I I
I I
I
II III I..
I
/ 0’6’0 ”’
I
I I
I. I
II.I
III II. I I
I» I I I I I .II. I.IIIII
II»II II II I» I III..}! II»
II II III IIIIIII‘
II II . IIII IIII I I
I II»IIIIII III . .III .» a.
.II. I.III I... .. 1... ......~ I» IIIII I
II 0 gt . ”IIIIIIII II9 I .I
I .I I» II I» IIII»I I»I III» “
l...II
o» I . IMIIII»I»IIII IINIIII»IIIIIIIII II 95% .
I » .I IIIII .II
I» I . 3562 III 33.3 II.III» IIIIIII I»II

 

 

      

 

 

   

 

 

       

 

II.I.»IIIHIMWE

rame areas.

Figure 14. Survey lines for which the surveyor listed white pine as the most abundant
timber type. Hatching represents present stump p

58

upland areas. Nevertheless, hemlock, white pine, and cedar were the species most often
cited first in these lines (Table 6). Spruce, yellow birch, and tamarack were often listed
second.

Surveyors also described the soil texture along each section line, probably based
primarily upon observations of the vegetation. The soils along almost every section line
in the study area were designated “sandy,” sometimes with the addition of “gravelly” or
“stony”. The soils for only five lines in the study area were described as “sandy loam”.
The surveyors also ranked soil quality. Only the terms second rate, poor second rate, or
third rate were used to describe the soils in the study area. Those lines where third rate
soil was noted have a good spatial correspondence with areas that today are stump prairie
(Figure 15), and lines where second rate soil was noted correspond well with certain areas
of current upland forest (Figure 16).

Judging from the surveyor line summaries, the spatial pattern created by the forest
communities of the pre-disturbance forest was apparent to the surveyors. A reasonably
experienced person asked to map forest types in the pre-disturbance forest, based on
species composition alone, might have drawn boundaries remarkably similar to those
defined today by the stump prairie/forest boundary, despite the lack of an obvious
geomorphic/topographic contrast. In fact, the plat map drawn by the deputy surveyor of
T 48 N, R 15 W in 1851 illustrates this. The portion of the county to the southwest of the
Kingston Plains is shown as having “Surface Level 2d rate soil sandy Timber Sugar
Maple, &c”, but the Kingston Plains area is labelled as “Surface gently rolling 3d rate soil
sandy Timber W Pine, Hemlock, Maple, Birch, &c.” The placement of the wording
suggests that the boundary between the two areas is approximately coincident with the
current forest/stump prairie boundary.

The line descriptions indicate that the surveyors recognized a distinction between
the forest vegetation present at that time in regions with contrasting vegetation types

today. In particular, white pine was present in areas that today are upland forest and

59

 

 

 

 

a»
~II III .II II

..IIIIIIIIIIuI .I.II
I~I

II

I...
I»

  

 

‘1
I.I~
gI
. I
.
...IIII
. .I.
. II. III. o .
o I I I I I
I 9” I ’0'
IIIIIIIIII IIIIuII» III IIIIII
.IIII I IIIIIII .
I.. I” IIIIIII
..II II III IIIIIIIII
”III-I II .Oo’o ’6’ II III” III’A.
..I...» .. ...... ......"
III I” III IIIIII IIII I

 

 

I
.II, ”I
II” I II o
I IIIIIII I .
I... I ”III/39%. av,
90/? I»I III I.
III.II.
. I. IIII.III.
o ,’ I
.II III .~ I.»
I I II I
IIIIIIII III IIIIIIIIIIIIII” I M. I I 3!?
IIIIII IIII IIIIII III/II II I.II/DI; a...» I
IIII I . I. II
IIIIIIIIIIIIIIIIIIIIIIIIIIIIII II I . ..II .~
VIIIIIIIII IIIIIII III IIII ..
.IIIIIIIIII III6IIII - l s
.II/II. . II
,
I I I I I I
a3.......x............ .... _ i
.2...,,,,.,;,,...:, .
III ”I!!!” It”!!! ’I”
III ”IIIIII IIIIIII III
I II II IIII I
II IIIII II .
III
IIIII ..
I
III I. ‘
II
I
I
. I
II.I
IIII I
”’O O I C O" 0
II I o
III.IIIII II» .II: “a”?
I. %
III .
IIII
5%W . 3550. III

            
 
                 

 

   
  

 

I
III III

.III I
sIIuawI 4%.... .II

 

         

    

I..)

II I
IQI w
I I I
IflIII I

  

III

 

 

 

             
          

 

 

  

 

 

  

   

  

 

 

 

 

 

 

 

 

 

 

 

 

 

rame areas.

Figure 15. Survey lines for which surveyor described the soil as “3d rate”. Hatching
represents present stump p

6O

 

 

     

 

   

 

 

 

Figure 16. Survey lines for which the surveyor described the soil as “2d rate”. Hatching

represents present upland forest areas.

61

swamp, but it was often the most prevalent species in areas that today are stump prairie.
Similarly, sugar maple, beech, and yellow birch were characteristically abundant in
current upland forest regions, but between the three were only once listed first by the

surveyors in places that have since become stump prairies.

62

Witness tree locations

The species, diameters, and locations of the witness trees recorded by the
surveyors also indicate a differentiation in pre-disturbance forest composition among the
current landcover types. Areas that today are upland forest contained significantly more
sugar maple, beech, and yellow birch trees than would have been expected from a random
distribution (see “Residual Score” columns in Table 7). Sugar maple was a particularly
distinctive component of this forest, as it was confined almost wholly to these sites (Table
7; Figure 17A). Yellow birch was occasionally encountered within areas that today are
stump prairie (Table 7), but most often was located at the edges of these polygons (Figure
17B). Beech trees, while present in the stump prairie areas, had a positive association
only with the upland forest polygons (Table 7; Figure 17C).

Sugar maple has been shown in many regions to occur only under particular
edaphic conditions. In the presettlement forests of Baraga County, Michigan, sugar
maple exhibited soil and site specificity, occurring primarily on mesic sites (Barrett et a1.
1995; Schaetzl and Brown n.d.). In the forests of northern Michigan today, sugar maple
also occurs only on better sites, and is absent on xeric sites (Shetron 1972). It was also
remarkably absent from ice-contact hills or outwash plains in northwestern Lower
Michigan (Host and Pregitzer 1992), the most xeric landforms. Thus, the absence of
sugar maple from the stump prairie sites may indicate that the stump prairie is more
edaphically xeric than upland forest areas.

In the current stump prairie polygons, only the two pine species had strongly
positive residual scores (Table 7). White pine, the most commonly recorded pine species,
was found throughout the study area, but was most frequent in stump prairie areas (Figure
18A). Red pine was less common than white pine, and was seldom noted outside stump
prairie sites. In fact, red pine witness trees were nearly confined to one group of stump
prairie polygons near the Lake Superior coastline (Figure 18B). Jack pine was noted

infrequently by the surveyors, and only in stump prairie areas (Figure 18C).

63

Table 7. Frequencies and standardized residuals expressing the degree of association
between species and current landcover classification.

 

 

 

Upland forest Stump prairie Swamp Other
Residual Residual Residual Residual

Species N score N score N score N score G2

Beech 305 55.7 65 -5.2 28 -52.9 0 -11.1 174.9
Cedar 23 -43.6 26 -2.8 112 115.6 5 0.0 156.0
Fir 76 1.5 30 0.1 22 -4.0 4 0.0 6.3”“
Hemlock 171 1.5 65 -0.1 66 -2.1 9 0.0 4.3*
Red Maple 99 6.5 35 0.1 17 ~12.3 2 -l .2 24.5
Spruce 35 -44.3 53 1.8 105 54.5 11 5.0 117.0
SugarMaple 202 71.6 1 -46.6 21 -22.7 0 -6.2 202.7
Tamarack 11 -65.2 16 -12.0 134 188.7 10 5.8 266.5
White Pine 107 -29.0 164 97.1 62 -9.1 24 20.0 160.4
YellowBirch 158 31.0 22 -11.0 23 -15.8 0 -5.6 77.2
Red Pine 1 -24.1 38 63.4 12 -0.1 1 -0.1 84.8

 

* P > 0.05. All other Gz values are highly significant at p < 0.0001.

{fink—.-

Hg'JIE 17
{01
BC

64

          

 

     

 
 
  

 

 

IIIIIII II. III III»I
:.I..... .... ......I......... ..........I..
.I.....I.. I. II... .I.. . .I.» I.
..I.. ..I.»..IIQI... .I.... .I ..... »..I... I.....

            

.I.

 
       

IIII
I

       

I
II»

      

         

    

      

       

     
  
  

 

                                      
       

    

    
    

        
 

 
 
 

  

 

II IIIII .III. I IIIIIIII IIII I I .
III.» I»»III . .II .IIIIII¢ II II .I ».y..... -I ..I
I? 9». IIII ...v IIIII ..III CIII.I IIII IIIIII/
. I. III I. I... .I. .I .II....I..r
I II ..III I. I III
I I II II III II... III
I I ‘III III a .I. III.. II I4 II
I»II IIII ..II. I I II. IIII I.I/QIIIII .III
IIIII»»I I‘ I .I .II I I.I»»IIII»II II II I.IIII. I»»I III
.I” I‘II»I III... I»» III .I.III II/ I.V.I.IV.IIIIIII III
II! IIIIII IIII IIII II»I IIIIII// .III. III
II. I 1... II.III I /I//. .I III
I . ..II... II.I..»». ..IIIWI». I. I... 1../I .I ..I.
I.III. ..IIIIII. I II.II II . III . II .I..
I.II I III .I.IIIII. I III II.III III .I I II
..I.. ...»...».... .I.III/..I.... .I.. ..I.. ..I.. ..I...» .I» II ..II
II II. I II I . I
I .III .I .III II»I I I
m». II.I... I»I»»II... .I. IIIfl»I III

  

I
.IIIIIII II
I... IIIIIII IlII II IIIII

 

    

 

       

     
 

   

      
        
  

     

 

    

                         

 
    
 

  
   

   

      

..I
.I.III..I.I. .I.II). II I. I
III . II II I II . II
. II.»»II.IoI1 .IIHIIIIIIIIIIIIIII»I I» I. I.I..aIIIIvIIIII.I1.... II .I.IIIIIII».. .I.wIIIIII» I II. I4..
.III... II». I»IIIIIIII ..III. ..I.. . .I .I» ..I III.. »II.... I.III»... III.. . .II. .‘I: I I... . ..II»...I ..IIIIIIII I.. .
II.IIIIIII.I IIIIIoIII I»IIII II’II¢I II IIIJtI tII»IIIIIIIIII~IIIII II.III II» III. II.I IIIdIII III .III a. .I.... I... II.II»I.I1II I»I... I. I II.III o
..I. II... ...» ..I.. II»..... I I..... ..III» I.. ..II/I.I. ..III .II I I. II... 11;? IX.
IIIIIIIIII/IIII .IIII . II I IIIIIIIIIIoIIII I.II». .III/III I I I’II‘»; .III .I ..III» I I.....»II II».»..II I..III. II.I II
II I I. I I I I II I». I..: .
II.IIIIIII I» III...» II.III”. »IIIIIIIII II.IIIIIIIIIII». ..II ..III .Ir II. .I.»... It.» I/II/II »I»......
I.IIIII: »I. IIIII»IIII.I . .I. II I.IIIIIIIIIII .IIIIII 1.5.9 I /I —.
II..»»..II». ..II ..II. II. .»II»»I§. II II Iw.‘ »II». ... I A 1%.... »
I.IIIIII I» IIIII»I I .II.I.IIIIIIIII.II»dII III .I III II /III .I.“/I— ” I»I.IIQIII II/I/IImIIIII I»I . I.....I II ..II»
»I I I II. III... .I.II II.I. I
II II1I..IIII I I I
..III »I ..IIIII »IIIIIII. .III. .I. I . I.IIII1/ II»..- . I I I». IfiII I...» t». w.
I.» .III I.. I.IIII»I II .III I I .I.II.IIIII» I I.IIIII»IIII I II »II
B I ..m»... ..I .I». I »‘II. I. I. I»»...II. WW
I.. II. III! I a - III.II9/I. IIZII II.II»/IIII.IIII.I III
I II I/IruI»I II». .I.II I. .I.III/I. II.III I.»IIII I
I I... .. II.III»I II.IIII I I I .II.III I ..I. ..I
I..! - ..II»».I .»I.»III 3...! II II.IIIIIII II.I-1a. .
.I .II II III.. II.IIIIIIIIII» III...I_ ..I..» ..III II»I. .713»...

 

III .I .I.II I I III. III II

I . . I
II..I.¢’I IIIII'I »IO IIIItI II.II II. I..!

    

 

     

     

I.IWWIIIII»

             

   

 

                  
                

  

   

    

   

 

   

 

   

  

 

               

 
 
  

I o I I III
I. . ..I. .II .I. I. I I. III... .III .I. I. .II. I
......III. ..I .I. . . .I....II. I.I II.II. I ..I .. I ......I...‘. I.».»».»»..~I..II...I....IIIII. III.....III I»» ..I....
......IIIIIIIII t.....». .I.... . ..I....IIIIIIIII»...I.I..I.I I.I...III . .I»-I. ......».....II»IIIIII..I..»..~r. I .I./.III... .I. ..I...
...-(.IIIIIono/IIIIII. I .I»»..IIIIIoflaItII I .4I7IIIIIII9 . ..IIIII . I I .I»» II.III»..IIII ..III..».. ..I. ......III». I». ..II
.I.II»...I I ....w... ..I. I..: ..I . .... .. .I . I..
I I . I I I II »II II.III »II I I. I .
I....I»II/I»I »III./...! ..I .I»...»I..II.III.II ..I... I... I. I. .. ... ..fl»..»»II.III»I ..III. ..I. I. .III. .»....../I.....I...~.II.. ....n......
I I . I- I II IIIII III I I .
II». I I II II III I... I III I.. ..II .II III
II I..».I I. III. I II II . .II I .II I...
II I I II II I .
..III I .III III. III I.I IIII I. I I.IIIIII I III/»IIIOIIII.’ IGIIMAI I I.. .
I »II.I 9 III I II» .II I I III III ... II
III.» II. III.. ..I.»I/» II.III/.I.III» . .. ..
.III I III ..I
II II.I 1 ..III.-
.III..».II»»II. 015.»: II I. II.I.I ...»»II I.IIIIIII. II . ..III.»
I II I II .I.. ... ..II. . .
............I. ..I............. I..... u. .. . ..
I I I .I.. I III I
I II
.IIIIII.I. I.I»..IIIII..» IWVIIIIIIIIQ. II I VIII.
IIIIIIII III II I.I I.IIIII II.I..I»
“I.III/.I.IIIIIII; .7¢.»1I.III
I»II.»»»II I.IIIIIIIIIII. II. III.....
.II..».I.... III.....IIII .I»I..I . . »I....
I I II.IIII’II I ..

          

II
III»I.IIII. .IIIIIIII II .IIIIIIIIII II.I IIIIIIIII
IIIIIII IhnIIamIII
»IIIIII IIIIIIII II II.I ‘I
I» ..I.. III..... I..» I.. I..»..III... .
II.III II.IIIII/IIu II. I II II.I...
I IfiIIIIZMIérII/II III1;II
,III I III
I. I. II
II III..... QIII I.III
IIIIIII.III III IIIIIIfiII .
III.
.I..

 

         
      

 
 

I
»J.» II.III»III.
I.I .

   

  

I.III I IIII .III.

Hatching represents present upland
B) Yellow birch witness trees. C)

165.

trees by spec

I

WlmeSS

ion of
forest areas. A) Sugar maple witness trees.

Beech witness trees.

17. Distribut

Figure

      

     

  
  

 

65

 

     
  
 
  

 

llwfll I‘llOIl l n .0. c : ’lu. -
”- llllIllloIlol l’llllllll - llflll llll llllll’ ”lul‘llloooll all lo-l? all
IIIMIIoll . lImllll l’Illl llIlllIlHII l IlIl ll//llll I ”-
.IIII I.II» I»IIII.IIII IIIII . II.III III II. . ».I II
I I/ .II I/IIIIII II I .. III
9I.I II . II» II :5.) II9».. II.I..II/I IIIII. .. II
..I... II. I. I. I.III». I..»IIII. .
III.. II .I. II. I
..III. I.» II.I IIIII .I.II... III/II.II I.III I I...
l on. o .
III.IIIII...I»I» I»II» II III.III I. II
IIII..I I I. III .III.IIII
III ”..I II III I I.II
..I... ..II/..III. ... . I
llllIIIIII I la ll l
I IIIIIIIIII II III” I I I.
......IIIIIIVIIII. I .
III
’H”” 0. use...
I I . . I...
’lol ll ll’l l o lo
.. .. .... ....
I I II .IIIIIIIIII” III“
I I ..II IIIIII I II
IO U

   

 

   

 

qol olll ll/U’lll ullllll

   

 

 

     

0 41 U lll
9.. o I ..III». .I.III»IIII..I.II».. I.. .II
I.
I I IIIOIIII‘/ IIII II I, I I
oI - II.III I“. 9. 4
I l I I'll llla .
... .IMIIIyI c ..wa.an
. .lloI o oolfl‘lll lIlll‘IIIlIIllI
I .....wII‘I I II.....IIIII.
I II... II III»IIIIIII .III. ..III IIIII..IIIII
I.I I IIIIIIIIII». IIIIII I IIIIII‘ IIII’IIIIIW
I.Inw‘I‘IwfiII’ . ¢HIII1I
flew-II.II? IIIII ..IIIIIIIII III. II
..III IIIIII
.. I.II...
IIIIIIIII .. a .
I II I. ..
...... .... . .. . . ..
I o I
o a u a. a . o a o o
.I u a o u u — o .
II. n | o I u u o a
o n
a
O o I
o
. ..I .
...
"..II I - IIIIIII .II
0’ llll-co . . I’Ol lo 0’ ll - ”I“
I I ...IIIIIIwII?»
. O I filllll

        
  

 

 

   

 

  

  
      

 

$6 ..I. I .II..III6 IMWI “.6.

  

 

 

 

I I .II.
.I I»» VIII III I». I....IIIIII .I»...»II‘ VIII»... I.II». L
III I II.III IIII IW/IIIIII/I II IIIII/I IIIII9
..I I. .I ..III I. ..I./I»III. II.II ..
.III. ..III. I. III .I.. I.IIIIIIII. I I ..I/II»IIII6I9 .. ..I III
I . .Il ofi/l-vI/o I ll l; u.
. I I.. I..
................. . ..I.... .
II I . .IIIIIIII. ”II.III. III.....II IvII. II .I.»IIII .I.III. . I .I.
. II.III .. I.. . . I. .
.II II .III IIIIIIIII .» I/..IIIIIIII».. .II.I.II . . .
I ..I... I.....»IIIIIIIIHII ..I I ..II... .. .I.
III II I III IIII II. IIII I ..
.IIIIIIIIIIIII III»IIIIIIIIIIIIIIIIIIIII I .III I.IIIIIIIIIIIIIIIIIII II II ..I I.I I
.I ..III I .9.» III.. I.I/III IIIIIIII.»IIII »IIIIIIII»IIII .I..
a II. .II .I.IIII.III.I II.IIIIIIIIIIIII IIIIIII.I.III»II .. .
I»I . .
IIIIIIII . .I.
lo 0
I
II I
II
I
I
.
ulola ll... I
.I»»I/II. I.II.II. I.. .I. . .I. II
lll o‘lll ‘l
I II.IIIIII» 4
..II I»I IIII II» .I
III... .... .. ..I....“ I. ..I»

 

 

    

     

   

 

 

Figure 18. Distribution of witness trees by species. Hatching represents present stump

'tness trees. C) Jack

mewr

tness trees. B) Red p'

te pine wi

i

.A)Wh

pine WlmCSS trees.

prairie areas

66

White pine is known to grow well on a wide variety of sites (Barnes and Wagner
1981; Schaetzl and Brown n.d.), but is found most often on the more xeric sandy soils,
especially better-developed Spodosols (Mokma and Vance 1989). In other parts of
northern Michigan, for example, pine species were predominant on drought-prone,
coarse-textured soils, sites with a tendency towards natural flammability (Whitney 1987;
Barrett et al. 1995; Schaetzl and Brown n.d.). In the presettlement forests of northern
Michigan white pine has been noted as an occasional associate in a wide variety of
community types (Hushen et al. 1966; Frederick et al. 1976; Whitney 1987; Barrett et al.
1995). Thus, the predominance of white pine on stump prairie sites could indicate that
these are xeric sites, even though scattered individuals of white pine also occurred in both
the upland forest and swamp forest regions.

Tamarack, cedar, and spruce were the species most often recorded by the
surveyors in swamp areas (Table 7). The strong association of these species with swamp
areas can be discerned from the species distribution maps (Figure 19), especially cedar
and tamarack. Spruce witness trees were less confined to swamp areas (Figure 19C),
probably because the surveyors did not distinguish between black spruce and white
spruce, which have different site preferences (Barnes and Wagner 1981). Spruce trees
occurring within stump prairie and upland forest may have been white spruce, while those
in swamp polygons were probably black spruce. Tamarack, cedar, and spruce trees were
also common in swamp sites in the presettlement forests of Baraga County, in the
western Upper Peninsula (Barrett et al. 1995).

Two evergreen species, balsam fir and hemlock, did not show a statistically
significant relationship with the current landcover classes (Table 7), and in fact were
scattered throughout the study area (Figure 20). The total number of hemlock witness
trees recorded was quite large; it was found on all four site types. Similarly, in
presettlement forests of Baraga County (Barrett et al. 1995; Schaetzl and Brown n.d.), as

well as in surrounding portions of Pictured Rocks National Lakeshore (Frederick et al.

67

 

        
     
     
              

   
     
  

 

 

   

 

          
         
                 

 

 

 

 

          

     
        

 

 

p
l :u- ... -uau m&
I. .vI. I I I I. . I I II....I. III III.. .I. Iv . I m a
...... . ......z. ......a... 2... w m
..OI. II‘ I III .I. I. .. .III. . I I/III.I
III I I IIIII II II ..I .I. I II.I I V I s p
I 0 II II ‘ II .- II II I I. I... II a CI
II I I ... I.. I. II I II ..IIIII / t S
.. ... . . 3...... a .. .. . ..I n
I I II . I....II e )
I” I.II I. .. IIIIIII S n v
no III. I.III. .- o I II.. R
. . D. S
I. .I.” B e
O . II n m
I
. c
.. .. . ..I A". .8» S
. . .II II 8
II. .. . ...I.. II ..I... r e
.... ..I . .. I». I Ito. III I I.. w m
I . I. I
I I. .I II IIanl/I/II ..I.. I.I. 4.33”. I....IIII III .I. .. I r .
cc 0 II II II I I oI - oI 0' .I I I o o
a. . ..I .M. II. ..VI 5.. . .II”. IunI ”3...... I g
If. III ‘I ’II“ III .I .. ”‘III IuI/Nfl II.I. I””””“ I I... ..I III” I.I.III. ”/IVII II.II ...II II II I III II. n
. no I.. I o o no . .
.II...I~ .I.III /I .II . ...... I . IIIIIIIIIIIgI. II.IIr. I? I. If I . I. .IIIo..II. “.m
In a ‘ . I .II .I .I IIIIII II” .I III . ... I. . .III. ...I I
III I o I. II.. .IIIIIII III III I A o . II .I I o. I In, I II c
IIII ’I II II IIIII I II III $4 II. IIII ..I II c
.. .I.. . .. . ..I: . .I...." 9.4.... ...??t... . . .. ... . . .. . ..I. m c
I .. . II ..I I. ..I III I I I .II . . . .
I I . II.I II II I I.. .. .II .. .I... . . a II. I a
. I . . . . III.... III II .I. III I . . I . I .I.. . .I .I.II H e
..I. I III II II III II.IIII III .I II . . I .I I
...... . .. .. . 3.2.1....” _ ...
. . I. . . II I II. 9 .. 0 III. I I. II . I.... I .
. . . III ..III ..I .III I II.II.I II . I o. I ..III. .III I . S
I .II..II II I II III..:{aIrISIIu . L.. l-LpII'sll: I .29. II I.. I ’~ ”II 1 _ e
B I. I III. I I..... I/III a I»: .1
. ....9 II. . . . ”III ..II“ c
. 5:. III II .II )
o. . . IfiaooI ”A. c I.. IIIIbflfllflu m B
u o o I
» .n.I 3.”qu ”I.I.IIH. I.. .I I. . w.”IIII ”awn“ S
. . . . I .. . . I. .IIIII I. o
- -- l ..I ... ”I. .2 ...... ......w ......I...”. v. S
v i .11 1 . .4 . a :11... .... ...‘I? . .uI. IIfi... ..I. ..I... ..I. .II t b C
4 4 1...: ... . .. In. .. . . I.. ..I. .I .. .I. .I. ... II.. . .. .../. .- .. ..I. C
...... .... s . . ...... .... .I.... ...... . .. . . . ... . ...... ...... .....xz s u
I I a I a .
..u....”I”I~IIIIHIII. .III ...II.‘ I.. I .III I. .../.II. .. II../“Inf.“ IwI/IIIIIIHHI I III/I II .IIIWII .I I. I”. .I.. .WIIInIIInaflnm” . a S
. . I . .. . III .II. .. .I I I. I I. I II II I
I . .. I III I .I ..... .I II.I / . II II I. I I II . . II II
3560 . .uu I.III. I.II InuIIIIIIIIIII. .. .I .III. . /I. I.. ..I I II. . ...IIIIIVI III.II. IIIII I III/IMIII - ...”.WHI I. I..”I’IIH ”IhwquIIIIIIMIIIIII” w a
. . I . . . . . I
II . ..I. I . . I/IflII .II: II. ..I . . ..I.I. I.II .... I: . I . I II V ..I. 1‘. II II.II I I A
III’. I II ..III II II .‘o I. I IIII , III I a II I I .I oI III IIII 4 IIIIIIIII 4 S
IIIIIII III./nII.IIII II I I” III I. . IhIIIIII . 0. IIIII .IIII. I V. II. II II.I VI” I.IIflIuIIIwIIIII/IIIflIIIIVIMOIIIIII I.II e .
I I . I I. . .I. I II. .II II.I II I I. IIIII ’ I .-
I II II II . II. . I I II I. p O I IIIII”I I I I I
I. .. . I . II. . I.» II I . ”II.II . I. III I III .. II .I. III. .
. o u a I I I III I IIIIIII o I
III . .../J: I »? ..I..u...III IIIIIII I.I.....mMIwIwiaz I9 . -.. - .-.... I M k
I.. . IIIIIII I . I I‘ I IE . c
I .I . ...HII 3:301»... I
II‘ I..II. .III/I I I .I. I .I I I‘I I. . .
.IIII. I . . ,I I .I I .. III » ..I. o
I III III’II II I I/I I I. III O IA
I I I o o. I I I I
I... ... . . .I.“? .I. . 9..” ”..I n
.IIIIWIII . .I I II IIhIII/II In...” IIIooIIII IIHIIINMv o m
III In I .I. III...“ I.. .II . I a IIIIIII t
I / I I II I. I I T
..H . .. ...... .. .... ...... u
I I I I I . I I
...... .. ., . ., ...... ...... .b )n
. II . ..I . I .I. . .I. . m
.I.III. Iflfl IINI’quIIflII .I.. .I .II3” II ’IHII. I.” H 8
II I II II Iv . .III I - I II.II II I III” I .1 o
.IAIIIIu .IIIIII // .III I ‘d I. IIII’IIII II a D
I I o I I I u
I III’IDIII‘II” II .*IIIIIIII . II .I.III.- .IIh III MMIII IIIIIII/I‘I ” o
. I II.II II I 4 II II. I I hIIII IIIIIIIIIII III II
o c I I II o o IMI III..-Io I III .I.III III‘ I I
I ..I. .II . ..III, III. I.I .III. I
I II I. o. I . II III .III... II I I
I... . .. II..... II I. II.I”IIIII/IHIIIIIIIIhIIIAIw... - .II...
A ..I.........n...,... .. t. .. g
o l

tness trees.

W1

68

  

 

                    
 

 

  

     

  

  
   

   

      

 

 

  

 

      

0 I III . h .I. III
I.I.. I ...»...ma».. IIIIIIIII.. IIIIIII..w.II:.
.. I I I I II.I...III ..I IIIIIIIIIuvIHIII I? .I.IIIIHIIII. I .
o I
I I ~II I ..IIIII.IIIII ”.I.III... IIII .I.II .M. IIII.IIIII”III III. . . a» II
I I I III I.
. I ..III I II‘. ..III. . IIIIII. I IIIIIV/IIIIIH II.II.II IIIIIIII I.. I IIIIP r
u I 9 I . II I I IIIIIII III II IIJ III. . .IIIIIIIIIII. j
... II II III 0 II.. .
. . II. I II‘III . . II‘IIIII IIIIIII II.II-I II III ”II.I IIIII"I/IIIII I C I IIII‘
5 II.. }. ..IMHIIIIXIII - I. . .. II.I... ...... {II.I .I. I. ..I ”I»III. I _
. IIIII .I.. I I I I I.
..I. II.II .II/IIIIII'IIIII. II II IuI IIIIII”..III.I . "I .II .. III I
I III. IIII IIIII . I I I
III ”III ”I II II.. .IIIIIIIIIIIII.IIIIIHI ..II II.II/I.IIIIIMIIHIIIHMWI II.III I n .I . . I
I.I..”..ul. I.. .I.III... .. I JIIW. I. I.IIIIIIIIonf .- I I I I I. I I.
I I’I III I IIIIII I C
II..”. I... IIIII4.I1..II.. II.III .. I I
II III-.IIIIIII II”...- I...I.II I III .III II I. I I.
O I I .
I I I
I II I I S I I‘ III
I I
II C O - I
‘. I I I
I u
I I I
I
”It I I I II II. IuIII/I .II‘ I
I -
IIII... I .II. I. .II.IIII I.—
I I I .I.IIIIIIII II”?
I . I II..”IIII IIIIII II~II¢ II III:
I I
IIIII .II .I I III II// II . w
I» i- o 1.............I.. .
I IIII;I. I.IIIOIIII.I ..III.IIIII .IIMIIII. I......II
II.III.III I.I II/.II IIIIII I/III IIIIIIfII III I
I ..IIII IIIflwIIIII/IMI IIII . I...I .IIIII IIII
I . . III I . .. I
IIIII II. . IIHIIII IIIIIIIIVIQ ..II. III IIIIIIII. ~ .I III
......I .../.... I.. .. .. . .....III .
III I I... I. I.III
II III .IMI III.. III II III I IIII... II.II . .
.I.IIIIIII»... II.III. III.....»IIIV I.I ..I IIIII.. II.III... . ....
. III I ”III I
.I..II ..I....III III..; ” /uIIII II I ..III I.
IIIII II. IIIIII IIII I..IIu .I/I. I.I .. I...
I .I.II/VIII... I... ..II. I I I ... t .
.. .II/III IIIII/III I, I I
I.IeII II.IIIII II.IIII .III.” I I . .
IIIII I I
.IIIIII IIIIIIIIIIII.III. I I
I .III IIII I O II.
.. ...... . ..
’ ..II 0
III III II
IIIIMII IIIII I I
II I I .
I .I
I III C .
I
I I
.II..- I I II. III .I .II¢.. I I I
III . I I.II .I. . I~ II.
II”. C III. IIGII I‘ h
I I I .II. I I I$~II~,
a IIII III I4
IIII C .III I II III” II . 1

    

 

 

. Distribution of witness trees by species. Hatching represents present stump

Figure 20

. A) Balsam fir witness trees. B) Hemlock witness trees.

prairie areas

69

1976), hemlock occurred across a wide variety of soil texture and drainage conditions.
Fir was less commonly recorded than hemlock in the study area (Table 7). Today it
occurs in Michigan on a variety of site types ranging from swamps to well-drained
uplands (Barnes and Wagner 1981).

Red maple was also recorded on all site types, although it was most strongly
positively associated with the upland forest polygons (Table 7; Figure 21). Red maple
thrives on a wide range of soil type and textures (Walters and Yawney 1990). In
presettlement Michigan forests, red maple has been noted for exhibiting wide site
tolerances (Hushen et al. 1966), although in Baraga County it was rarely recorded by the
surveyors (Barrett et a1. 1995). It is a common species today in the study area both as a

member of the forest canopy and as isolated individuals on the stump prairie (Table 2).

7O

  

   
          
    
           
    

 

0

0

II.

m....II

 

 

. C ' I .IIII .IIIII I ”II.III IIIIIIIII ”fig/II I o ’I
0‘ III ow- .III - IIIIII o I. 'I’
II.. ..I I II LI. ”II.I." ..I I. I..III/IIIIIIII .. I.
5.0“ .III I’oI III .I.I- I I III I”.
. I I.I II IIII. I I. II. II III .IIIIII
. »III .II?” I I on IIIIIII IIIIIII.IIIII IIII IIII.I II I II III.II I I.
IIIIII II I IIIIII III.III II ‘ II III.... I II
II ... I.. quIII/II - .III I I35 IIIIII ..IIIII/II/I. l. ..II/»III.?
II II I I I .
...I. ”.I.II.I IIIIIIIIIHIIIIIIII IIIIII.I II. .IIIIIIII IIIIII InnIII/I/ Iqu a II I .I.II o
..IIIIIIII ”IIIuIIII... . IIIIIIIIII.IIIII o I .IIIII IIIIIII I III.IIII I o o .o o I
II I IIIIHI II.I IIIIIIIIII I III.I .. I.II/III IIIIIIIII I I a II a I
I I II II IIIIIIII I III I III IIIII I II I I I I
III IIII I.I IIIIII- III IIIIIQI IIIII III 5 l a . o-mIII .
IIIIIII: .III o/IIIIII IIIIIII IIIIIIII/ .
. I II I ..IIIII IIIIIIIII - o .
III IflI II.. I.III. .I II III . o. . 0..
II II ..IIIIIIIIII IIIIIIIIII III I.IIo - a on o - ..
IIIIIII IIII o. .
- IIIIIII IIIIIIII o
o IIHI . I . u I . U
I O
I II I D
II. a I u o g o o
a o o o
”I.. I I a II. INIIIIII . a... .
IIIIIII III II. .II I It‘IWflIIII am“
I
I O o IIIII
I I ..II IIIII.III~ II II
III.
IIIII . ..III PI III. .III; IIIWII W.

 

 

 

witness trees. Hatching represents present stump

Figure 21. Distribution of red maple

prairie areas.

71

Witness tree diameters

When the diameters of the witness trees and the species distributions are taken
into account, the distinctions among the forest communities on the four landcover types
become more muted (Table 8). In all four landcover classes, white pine has the highest
modified importance value, mostly due to its large basal area. To a lesser extent,
hemlock also shows a high modified importance value in all landcover classes due to its
larger basal area.

Notable differences among the forests of the landcover types are, however, still
evident. The importance of white pine in the stump prairie areas (48.6) is proportionally
much greater than its importance in either the upland forest (20.0) or swamp areas (20.6;
Table 8). The distinction between the upland forest and stump prairie areas is most
evident when one examines the data for yellow birch and sugar maple. Both species have
high importance values in upland forest polygons (yellow birch, 15.1; sugar maple, 14.3)
due to high relative density and relative dominance. Similarly, both have quite low
importance values in stump prairie areas (yellow birch, 3.6; sugar maple, 0.2) (Table 8).

The size class distributions of the white pine tree diameters (Figures 22A and B)
illustrate the extraordinarily large diameters of the white pine trees, especially those
located within upland forest polygons. The broad size distribution of the white pines in
the upland forest areas suggests that the trees were of uneven ages, and did not reproduce
solely in response to widespread disturbance. The white pine trees in the stump prairie
areas also show an uneven age distribution, with the exception possibly of a large cohort
in the 25 - 30 inch diameter class.

Beech trees in the upland forest areas have an inverted “J” size class distribution,
typical of a reproducing shade-tolerant tree (Quinby 1991) (Figure 23). In the areas that

are today stump prairie, the beech trees were especially small, with no tree recorded with

72

Table 8. Relative density, relative dominance, and modified importance values of
witness trees grouped by current landuse classification.

 

 

 

 

 

 

Total Mod.
Mean Max Total basal Rel. Rel. Import.
Species N DBH DBH DBH are‘a Density Danimne Value
2111 m‘ %2
Upland forest:
Whitepine 107 65.0 152.4 6946.9 44.91 8.9 31.0 20.0
Beach 305 25.9 63.5 7894.3 18.19 25.5 12.5 19.0
Hemlock 171 39.6 203.2 6766.6 26.57 14.3 18.3 16.3
Yellowbimh 158 39.6 203.2 6263.6 24.61 13.2 17.0 15.1
Stmnnple 202 28.7 200.7 5773.4 17.07 16.9 11.8 14.3
Redmaple 99 26.2 45.7 2588.3 5.90 8.3 4.1 6.2
Balsanfir 76 22.6 35.6 1714.5 3.24 6.3 2.2 4.3
Spruce 35 24.6 38.1 861.1 1.77 2.9 1.2 2.1
WhiteCedar 23 27.4 50.8 629.9 1.47 1.9 1.0 1.5
Tamarack 11 25.4 45.7 279.4 0.62 0.9 0.4 0.7
Amman 2 40.6 40.6 81.3 0.26 0.2 0.2 0.2
Whiebildi 2 38.1 45.7 76.2 0.24 0.2 0.2 0.2
Bladcamy 2 16.5 17.8 33.0 0.04 0.2 0.0 0.1
Main 2 12.7 12.7 25.4 0.03 0.2 0.0 0.1
Redpine 1 25.4 25.4 25.4 0.05 0.1 0.0 0.1
L 20.3 20.3 20.3 0.03 0.1 0.0 0.1
Total: 1197 33.3 203.2
Stump prairie:
thpine 164 53.3 106.7 8757.9 44.20 31.2 62.4 46.8
Hemlock 65 35.1 91.4 2280.9 7.48 12.4 10.6 11.5
Beach 65 24.1 35.6 1562.1 3.09 12.4 4.4 8.4
Redpine 38 43.2 86.4 1643.4 6.33 7.2 8.9 8.1
Spruce 53 21.3 33.0 1125.2 2.01 10.1 2.8 6.5
Redmzple 35 23.1 30.5 807.7 1.56 6.7 2.2 4.4
dewbild‘l 22 34.0 61.0 749.3 2.19 4.2 3.1 3.6
WhiteCedar 26 26.2 45.7 678.2 1.56 5.0 2.2 3.6
Balsmfr 30 19.1 27.9 574.0 0.91 5.7 1.3 3.5
Tanaradc 16 23.1 40.6 368.3 0.75 3.0 1.1 2.1
Jackpine 7 26.4 35.6 185.4 0.45 1.3 0.6 1.0
Whitebimh 3 25.4 35.6 76.2 0.16 0.6 0.2 0.4
Sugarmple 1 38.1 38.41 38.1 0.11 0.2 0.2 0.2
Total: 525 35.8 106.7

 

73

Table 8. (Cont.)

 

 

 

 

Total Mod.
Mean Max Total basal Rel. Rel. Import.
Species N DBH DBH DBH area Density Domimnoe Value
------------ cm------------ 111‘ °A2

Swamp:
Wpine 62 47.8 127.0 2954.0 13.78 10.1 31.2 20.6
WhiteCedar 112 25.1 63.5 2804.2 6.23 18.3 14.1 16.2
Tm 134 18.5 40.6 2491.7 4.32 21.9 9.8 15.8
Hemlock 66 35.6 76.2 2347.0 7.45 10.8 16.9 13.8
Spruce 105 19.8 40.6 2080.3 3.60 17.1 8.1 12.6
Yellowbimh 23 31.2 55.9 716.3 1.93 3.8 4.4 4.1
Beach 28 24.4 40.6 685.8 1.43 4.6 3.2 3.9
Sugrmaple 21 30.5 50.8 640.1 1.77 3.4 4.0 3.7
Redpine 12 36.3 66.0 434.3 1.44 2.0 3.3 2.6
Bahamfir 22 19.1 30.5 419.1 0.68 3.6 1.5 2.6
Redmmle 17 25.1 40.6 426.7 0.92 2.8 2.1 2.4
thkash 4 26.2 30.5 104.1 0.22 0.7 0.5 0.6
A3331 2 26.7 30.5 53.3 0.11 0.3 0.3 0.3
Whitebirdi 2 20.3 30.5 40.6 0.08 0.3 0.2 0.3
Blad<Chary 1 40.6 40.6 40.6 0.13 0.2 0.3 0.2
Bird: 1 35.6 35.6 35.6 0.10 0.2 0.2 0.2
Jad<pine 1 15.2 15.2 15.2 0.02 0.2 0.0 0.1

Total: 613 26.7 127.0

Other:
Whitepine 24 40.6 76.2 972.8 3.48 36.4 63.6 50.0
Hemlock 9 29.7 50.8 266.7 0.68 13.6 12.5 13.1
Spuoe 11 21.1 30.5 231.1 0.42 16.7 7.7 12.2
Tamaad< 10 22.9 35.6 228.6 0.45 15.2 8.2 11.7
WhiteCech' 5 23.4 30.5 116.8 0.22 7.6 4.0 5.8
Balsmfir 17.3 20.3 68.6 0.09 6.1 1.7 3.9
Redmple 20.3 30.5 40.6 0.08 3.0 1.5 2.3
Redpine 25.4 25.4 25.4 0.05 1.5 0.9 1.2

Total: 29.5 76.2

 

 

74

 

 

7O

 

 

 

 

5 (12.7) 15 (38.1) 25 (63.5) 35 (88.9) 45 (114.3) 55 (139.7) More
Dianeter 1n inches (an)

 

 

 

70

 

PM
8

 

 

 

 

 

5 (12.7) 15 (38.1) 25 (83.5) 35 (88.9) 45 (114.3) 55 (139.7) More
Diameter In Inches (cm)

 

areas

Figure 22. Diameter size class distributions from GLO notes for white pine trees in A)

currently mapped as upland forest and B) areas currently mapped as stump

prairie.

 

 

75

 

 

 

 

J J

5 (12.7) 15 (38.1) 25 (83.5) 35 (88.9) 45 (114.3) 55 (139.7) More
Diameter In Inches (cm)

 

 

 

 

Figure 23. Diameter size class distributions from GLO notes for beech trees in areas
currently mapped as upland forest.

76

a diameter larger than 15 inches. This may indicate that beech reproduction on the stump
prairie was of relatively recent origin at the time of the survey, perhaps due to a
disturbance that had occurred previously, killing the beech trees. The distribution of
sugar maple diameters in the upland forest polygons is similar to that of the beech trees

(Figure 24).

77

 

7O

 

 

  

L

 

 

5 (127) 15 (38.1) 25 (83.5) 35 (88.9) 45 (114.3) 55 (139.7) More
DIemeter In Inches (an)

 

 

 

Figure 24. Diameter size class distributions from GLO notes for sugar maple trees in
areas currently mapped as upland forest.

78

Stumps

The stumps from nineteenth century logging are the last remaining physical
evidence of the character of the pre-disturbance forest. The evidence available from the
stumps must be interpreted with caution because the effects of a century of decay on the
stumps are not known. On the stump prairies, many stumps appear to be well-preserved
as to the original placement and dimensions of the stumps, although the interiors of many
of the stumps are rotting (Figure 25). Some stumps are in a more advanced state of
decay, and are thus more difficult to measure. Bark is lacking on all stumps, but exterior
rings of wood on stumps of the open prairies are, in most cases, intact. In the forests,
however, the moister microenvironment has often apparently hastened the decay process
and the exterior wood of the stumps is usually rotting. In the forests there is also an
additional sampling problem: one must distinguish between nineteenth century stumps
and stumps from more recent logging efforts. The presence of bark on a stump was taken
as evidence that it was not from the original logging. Underbrush and recent logging
slash also occasionally made finding stumps in the forest difficult.

It is likely that all, or nearly all, of the stumps remaining from the nineteenth
century represent white pine or red pine trees. Wood from five stump prairie stumps, one
in good condition and four relatively decomposed, was identified to the genus level. All
five are from the genus Pinus. Thus, the level of decomposition cannot be taken to
indicate a difference in the stump’s species. Stumps of hardwood trees have probably
completely rotted away.

Deming. On the average, there are slightly fewer stumps per hectare in the
forested stands (91.8) than in the stump prairie areas (113.2; Table 9). This can be

attributed to two factors:

79

 

Figure 25. A representative view of stumps on the stump prairie.

80

Table 9. Diameter at breast height (DBH)l and absolute density for logging era stumps in
sampled stands.

 

 

 

Maximum Absolute
Stand Mean DBH DBH Total DBH Total basal density
area
m m‘ 5113111211131

FOR-1 54.4 100 1794 9.13 79.2
FOR-2 ...Z
FOR-3 53.1 110 2070 10.00 129.5
FOR-4 60.3 100 2230 1 1.38 72.0
FOR-5 59.5 100 2380 12.00 96.5
FOR-6 47.3 90 2270 9.15 221.9
F OR-7 67.3 120 2220 12.96 55.3
FOR-8 69.7 120 2580 15.80 76.8
FOR-9 52.1 100 2190 10.16 86.8
Total: 57.4 120 91.8
SAV—l 48.0 90 2830 11.80 88.7
SAV-2 45.6 90 2370 9.57 205.0
SAV-3 52.1 90 2500 11.58 127.2
SAV-4 50.9 110 2850 12.39 113.3
SAV—S 58.7 90 3520 17.40 159.6
SAV-6 57.8 120 3180 16.41 73.3
SAV-7 55.4 110 3105 15.50 101.8
SAV—8 51.5 110 2470 11.87 107.1
Total: 52.6 120 113.2

 

 

 

IEstimated diameter at breast height of tree based on diameter of top of stump.

2Stumps were not sampled in stand FOR-2 due to the large amount of recent logging

slash in the stand.

125:
511111

Slim

mear

1113311

81

(1) The stumps remaining from the nineteenth century logging are

primarily the stumps of white pine trees and thus represent only a partial

record of the pre-disturbance forest. Based on the GLO notes, white pine

represented a smaller proportion of trees in areas where the forest has

regenerated than in present stump prairies (Table 8). If the stumps of other

tree species have preferentially decayed, the density of the remaining

stumps should be lower in the forest, where formerly there had been more

trees of other species.

(2) Stumps in the forest are difficult to find and are more likely to decay

than are stumps in the stump prairies, as discussed above. Thus, it is

possible that some stumps in the forest went unsampled or have

disappeared due to decay.

Qiamflem; The average diameters of stumps in the forested stands (5 7.4 cm) are
slightly larger than the diameters of stumps in the stump prairies (52.6 cm; Table 9). In
forest and stump prairie stands located near each other, the stumps in the forested stands
are larger than the stumps in the adjacent stump prairie (Figure 26). Given the
assumption that most stumps represent white pine trees (see above), this pattern would be
expected if the white pine trees in the forested areas were scattered, emergent giants,
while the forest in the stump prairie regions was dominated by white pine.

Alternatively, larger remaining stumps in the forested areas could be attributed to
faster decay rates in the moister environment of the forest. It is possible that smaller
stumps in the forested stands have decayed and disappeared, leaving only the larger
stumps to be sampled, or that they were overlooked in sampling, as discussed above.

Mean diameters of the stumps in the sampled stands are roughly equivalent to
mean diameters of white pine trees as recorded by the GLO surveyors. Average
diameters of white pine trees from the GLO notes in regions that are now upland forest

(25.6 inches or 65.0 cm) was slightly greater than diameters from the GLO notes of white

\\

. \\
96% DH qfimv
a.

\

\
. \\
\ \\\\\ .

\
\\

\

s
5

\§ A
\\ NQ®
\ m...

c\ ....

..v

  
 

83

pine trees in current stump prairie areas (21.0 inches or 53.3 cm; Table 8). Both of these
values are slightly larger than the mean diameters of the stumps in the sampled stands
(forest stands, 57.4 cm; stump prairie stands, 52.6 cm; Table 9), but are comparable
considering the loss of bark (and possibly some outer wood layers) from the stumps.

The size class distribution of the stump diameters (Figures 27A and B) illustrates
the slightly larger mean diameters of the stumps in the forested stands. Both distributions
are broad, suggesting that the trees they represent were of uneven ages, and did not
reproduce solely as a response to widespread disturbance. The size class distributions of
the stumps are comparable to the distributions of white pine tree diameters in the pre-
disturbance forest as indicated by the GLO surveyor’s notes (Figures 22A and B ), which
also showed broad diameter distributions in both current landcover types.

Information from stump locations and dimensions appears to support inferences
about presettlement forests derived from the GLO notes. Specific conclusions about the
nature of the forests are difficult to make on the basis of the stump data alone due to
uncertainties about preferential decay and decay rates. Nevertheless, the stump data do
suggest that the forests that existed on the sampled sites were similar to the forests
reconstructed from the GLO notes of the region as a whole in the two contrasting current

landcover classes.

84

 

7O

 

Percent
8

10 ~~

 

 

 

 

 

o'm'e'n 8‘9 8 :2 :2 a a :3
Dluneteucm)

115

25

135 ‘
Mom ..

 

 

 

 

70

 

 

 

 

 

 

{We're 149 :3 e e a s:
DIemeter(cm)

105
115
125 ‘
135 ‘
Mon

 

 

 

Figure 27 . Diameter size class distributions for stumps in A) forest stands and B) stump
prairie stands.

85

Comparisons with current stand composition

A comparison of the present species composition of the sampled forest and stump
prairie stands (Table 2) with the pre-disturbance species composition of these areas as
reconstructed from the GLO notes (Table 8) illustrates the changes that have taken place
presumably due to repeated disturbance (Table 10). The disturbances affecting the
species composition of the study area include both the original nineteenth century logging
and associated fires, as well as more recent timber harvesting and land management
practices.

In regions that are currently forested, three of the four most common species
today (white pine, beech, and hemlock) were also the species with the highest modified
importance values in the GLO surveyor’s notes for the same regions (Table 10). Red
maple, which has the highest importance value (38.4) in the sampled stands, however,
ranked sixth in modified importance value in the GLO notes, behind yellow birch and
sugar maple (Table 10).

Neither yellow birch nor sugar maple were encountered in any of the sampled
forested stands. I have seldom seen either species in any of the forests adjacent to the
stump prairie boundaries, except for some sugar maple in the extreme northeastern comer
of the study area (sections 2 and 3 of T 48 N, R 15 W) and also in the southwestern
comer of the study area (section 30 of T 48 N, R 15 W). A few yellow birch trees have
been observed near Wise Lake in section 27 and also in the southwestern comer of the
study area (section 30). Although sugar maple and yellow birch were both prevalent
species in the pre-disturbance forest in regions that are today upland forest, sugar maple
was not recorded by the surveyors in the upland forests immediately adjacent to the
current stump prairie study area (Figure 17A). Yellow birch was found nearer to the
stump prairie areas (Figure 17B), yet it was most common in other parts of the upland
forest as well. Thus, the importance of sugar maple and yellow birch in the presettlement

forests of areas currently mapped as upland forest was probably greater farther away from

86

Table 10. Importance values and ranks of tree species as compiled from GLO notes and
from modern sampled stands.

 

 

 

GLO notes Modern sampled stands
Species Mod. Imp. Rank Imp. Val. Rank Change in
Val. Imp. Val.

Forest:
Balsam fir 4.3 7 9.2 5 +4.9
Beech 19.0 2 17.2 3 -1.8
Hemlock 16.3 3 9.9 4 -6.4
Red maple 6.2 6 38.4 1 +322
Red pine 0.1 >8 2.9 7 +2.8
Spruce 2.1 8 0.3 >8 -1.8
Sugar maple 14.3 5 absent absent -14.3
White birch 0.2 >8 4.2 6 +4.0
White pine 20.0 1 17 .3 2 -2.7
Yellow birch 15.1 4 absent absent -15.1

Stump Prairie:
Balsam fir 3.5 >8 0.2 8' -3.3
Beech 8.4 3 absent absent -8.4
Black cherry absent absent 5.0 3 +5.0
Black spruce 6.5 5 2.5 6 -4.0
White cedar 3.6 8 absent absent -3.6
Hemlock 1 1.5 2 absent absent -1 1.5
Jack pine 1.0 >8 3.9 4 +2.9
Juneberry absent absent 3.5 5 +3.5
Red maple 4.4 6 8.7 2 +4.3
Red pine 8.1 4 2.3 7 -5.8
Tamarack 2. 1 >8 absent absent -2.1
White pine 46.8 1 47.8 1 +1.0
Yellow birch 3.6 7 absent absent -3.6

 

lQuadrants of point-quarter sampling in which no tree was recorded due to location > 30
m from the sampling point (“missing” trees) have been omitted in the ranking in
this table.

87

the current stump prairie than in places immediately adjacent to it, where the current
stands were sampled. Therefore, this apparent change in forest composition may be an
artifact of different geographical areas of sampling (i.e., six survey townships vs. stands
immediately adjacent to the stump prairie border) rather than a real shift away from sugar
maple and yellow birch in the sampled stands. The fact that both sugar maple and yellow
birch are common today in some nearby upland forests tends to support this
interpretation.

In areas that are today stump prairie, white pine was the species with the highest
modified importance value (46.8) in the pre-disturbance forest; it remains the
predominant species in the stump prairie stands (importance value = 47.8; Table 10).
Hemlock and beech, the second- and third-ranked species from the GLO notes, were not
encountered in any stump prairie sampled stand. The next-ranked species from the GLO
notes, red pine and spruce, have relatively low importance values in the present sampled
stands (red pine, 2.3; black spruce, 2.5; white spruce, 1.5). Red maple, however, has
become proportionately more important following the disturbance, and is now the
second-ranked tree species in terms of importance value. Other common trees of the
stump prairie today, such as black cherry and juneberry, were not recorded by the
surveyors.

The absence of hemlock and beech from the stump prairie is probably due to the
reproduction requirements of the species. Hemlock is characteristic of cool, moist, highly
acid sites and typically requires moist, cool, shady sites to establish (Barnes and Wagner
1981). If the hemlock trees in the stump prairie areas were killed by fires or logging at
the end of the last century, the heat and drought that prevail on the deforested stump
prairie might have prevented its re-establishment. Likewise, mature beech trees are
highly susceptible to fire, and therefore were unlikely to have survived the fires which
followed logging. Beech seeds are produced in large numbers only at infrequent

intervals, and much beech reproduction is from root sprouts (Curtis 1971). These

C111

0111

88

characteristics make re-establishment following the logging and burning which took place
on the stump prairie unlikely.

Red maple has increased in prominence in both the forested and stump prairie
regions (Table 10). In forested areas, it ranked sixth in the pre-disturbance forest, but
today it ranks first (Table 10). In stump prairie stands today, it ranks second only to
white pine (Table 10). The increase in prominence is probably due to the species’ ability
to grow in a wide range of site conditions (Barnes and Wagner 1981) and to the fact that
it sprouts vigorously following injury by fire, cutting, or browsing (Walters and Yawney
1990). Thus those few trees that were present in the pre-disturbance forest were able to
reproduce following logging and fires, and survive the harsh, dry conditions that
prevailed on the post-disturbance stump prairie. White birch has also increased in
prominence in the forested sites, though not to the degree of red maple (Table 10),
probably because it, too, can regenerate from sprouts following cutting or fire (Safford et
al. 1990).

In stump prairie stands, jack pine has increased in prominence (Table 10). Jack
pine is a fire-dependent species and is adapted to reproducing following fires (Barnes and
Wagner 1981), so the increase in prominence may be due to increased jack pine
reproduction following late nineteenth century fires. The increase in prominence may
also be associated with widespread planting of jack pine in the study area. Jack pine
appeared to be more common in those stands that were near jack pine plantations, and
some of the jack pine trees encountered in sampling may be descended from the planted

trees.

Summary
In summary, evidence from the General Land Office survey notes suggests that
considerably different forest types existed in the areas that have reforested after the

logging of the last century, when compared to those that have remained in stumps. In

89

particular, the forests in the stump prairie areas were predominantly pine forests, with
major components of white and red pine. Hemlock and hardwood species did exist in
these areas, but in comparison to their role in the surrounding forest types they were
relatively unimportant. Especially notable is the near lack of any record of sugar maple
trees in the areas that today remain stump prairies, and its prevalence in the GLO notes in
areas currently mapped as upland forests. Forests in current upland forest areas were
primarily beech-sugar maple-yellow birch forests, but also contained some white pine and
hemlock. Today sugar maple and yellow birch are absent from the forested stands
immediately adjacent to the stump prairie, and the forests are predominantly red maple-
white pine-beech forests. Red maple has especially increased in prominence in these
forests.

The pre-disturbance forest of the stump prairie regions does not appear to have
been a particularly impoverished forest community, or a forest grading towards a
savanna. It was dominated by large white pine trees, with a sizable admixture of hemlock
and hardwoods such as beech and yellow birch. Rather, the contrast between the forests
of the stump prairie and upland forest regions appears to have been more of species
composition than of density or basal area per ha (dominance). Today the stump prairie
stands have only scattered trees, and the most common of those include white pine, red
maple, and black cherry. White pine has continued to be the first ranked tree in the stump
prairie stands.

Due to the high proportion of white pine in the forest, the stump prairie areas were
the most attractive sites for lumbering in the late nineteenth century. Where white pine
was most common, the true “pineries”, logging and subsequent fires had preferentially
removed the forest vegetation by the end of the nineteenth century, leaving large tracts of
bare land “where the ground is sparsely covered with weeds and grass, sweet fern, and a
few scattering rtmty bushes of scrub oak, aspen, and white birch” (Roth 1898, p. 13). In

the hardwood-dominated areas, however, the removal of scattered giant pines was less

90

noticeable, and, according to contemporary accounts, fire damage to the forest was

minimal except where it bordered on pine slashings (Roth 1898, p. 12).

91

Factor 2: Substrate and soil properties

An understanding of soil and its parent material is important in a study of an
ecosystem, because the soil and the vegetation are interrelated components of a single
system, and thus influence each other. This chapter is divided into two parts. In the first,
soil and substrate properties and their relationship to, and possible influences on, the
spatial patterns of forest and stump prairie in the study area are discussed. The second
part examines the soil data, including information from both the solid phase of the soil
and from buried resin bags, for insight into the active processes of soil development in

the forest and stump prairie areas.

Substrate, soil, and current forest patterns
E . l . l E

Although the stump prairie/forest boundary does not follow any obvious
topographic or geomorphological boundary, it is possible that the stump prairie and forest
regions have different depositional and sedimentary histories, and, therefore, that the
parent materials of the soils are different. Because of the extremely sandy textures of the
soils in the study area, slight variations in parent material silt and clay content could
affect forest regeneration by controlling water holding capacity. This section will discuss
the relationship between the texture of the parent materials and spatial patterns of forest
regeneration.

Wm. A comparison of soil C horizon particle size
distributions between pedons in the forested region and those under stump prairie shows
that there is no significant difference between the two groups in terms of sand, silt, or
clay content (Table 11). The C horizons of all sampled pedons are almost devoid of silt

and clay.

92

Table 11. Uppermost C horizon particle size distributions for sampled pedons.

 

 

 

Sand Silt Clay vcs cs MS is vrs

2.0-0.05 0.050.0m <0.002 2.0-1.0 1.0-0.5 0.50.25 0.25.01 0.10.05
Pedon mm mm mm mm mm mm mm mm

------ Wilm- W911
FOR-1 99.6 0.2 0.2 1.7 16.7 61.4 19.3 0.8
FOR-2 99.5 0.1 0.4 3.5 15.7 70.1 10.3 0.3
FOR-3 99.7 0.1 0.2 0.2 4.4 71.5 23.3 0.7
FOR-4 99.5 0.3 0.2 0.4 8.5 51.0 38.4 1.8
FOR-5 99.6 0.1 0.2 0.2 10.7 77.3 11.6 0.3
FOR-6 98.0 1.7 0.3 5.2 16.2 48.8 27.1 2.7
FOR-7 99.6 0.2 0.2 1.2 20.4 68.4 9.7 0.3
FOR-8 99.5 0.3 0.2 2.7 18.3 61.5 16.9 0.7
FOR-9 99.6 0.2 0.2 0.7 12.7 60.1 25.5 1.0
Mean] 99.4 0.4 0.2 1.7 13.7 63.3 20.2 0.9
SAV-l 99.3 0.2 0.5 2.1 7.9 51.1 37.3 1.6
SAV-2 99.8 0.1 0.1 0.2 5.1 81.1 13.3 0.3
SAV-3 99.6 0.1 0.4 0.0 0.3 37.9 60.3 1.4
SAV-4 99.8 0.1 0.1 0.8 14.4 77.2 7.4 0.2
SAV-S 99.8 0.1 0.1 1.8 5.7 59.7 31.2 1.6
SAV-6 99.5 0.3 0.2 0.0 1.1 55.4 42.0 1.4
SAV—7 99.7 0.2 0.1 2.0 11.6 64.8 21.2 0.4

SAV-8 99.4 0.2 0.4 9.3 38.2 43.4 8.6 0.5
Mean‘ 99.6 0.2 -0.2 2.0 10.5 58.8 27.7 0.9

 

 

TMeans are not significantly different at p < 0.05 (Kolmogorov-Smirnov) for all columns.

93

In many pedons, faint banding or stratification of sands was observed below the
solum (Table 12). Clay content increases, as determined in the field, were minimal in
most of the bands. Many bands were apparently sedimentary concentrations of darker
colored or iron-rich sands that were continous around the entire pit. Most were too thin
for separate sampling or color determination. Bulk samples from the affected depths
show no clay content increase over the upper C horizon (Table 22 in Appendix A).
Similar banding and faint lamellae occurred in both forested and stump prairie pedons.

Three forested pedons (FOR-6, F OR—7, FOR-9) do contain sedimentary lenses of
finer, pinker sands. In the case of FOR-6, an increase in roots was observed in the
stratum (sampled as the 2C horizon). The stratum contains little clay, but almost 9% silt,
as well as more fine sand than the surrounding horizons (Table 22 in Appendix A). The
pinker stratum in FOR-7 has more clay (2.9%) and silt (2.0%) than the adjacent C
horizon (clay 0.2%; silt, 0.2%; Table 22 in Appendix A). In FOR-9, the stratum occurs
well below the solum (270 - 290 cm), and has a notable increase in silt, fine sand, and
very fine sand over adjacent horizons (Table 22 in Appendix A). These sedimentary
strata may have had an effect on tree growth by increasing the 5011’s water holding
capacity, especially in pedon F OR-6 where tree roots were observed in the stratum.
Nonpedogenic textural bands have been shown to affect forest composition (Host et al.
1988; Host and Pregitzer 1992) and site index (Hannah and Zahner 1970), especially
when they occur within 2 m of the soil surface. Nevertheless, not all forest pedons
examined for the present study have such sedimentary strata within the upper three m,
and some stump prairie pedons do. Thus, deep textural bands and finer-textured
sedimentary strata do not appear to be directly related to the recovery of the forest from
the nineteenth century logging disturbances.

Wm. For all pedons, weighted solum means show that
more than 96% of the fine earth fraction falls into the sand size class (Table 13). This
percentage is slightly lower than that of the C horizon alone (98%, Table 11), probably

94

Table 12. Banding and lamellae encountered below the solum in sampled pedons.

 

 

Pedon Depth Description
12m

FOR-1 140 - 200 Few lamellae, 1 mm thick, 7.5 YR 6/4.

FOR-2 200 Few very fine lamellae.

FOR-3 74 - 150 Thin dark strata, some orange strata, up to 2 mm thick, 1 - 2 cm
apart.

F OR-4 none

F OR-S none

FOR-6 98-108 Sedimentary band 10 cm thick, finer sand, some field-detectable
clay increase, 7.5 YR 5/4.

FOR-6 108-130 Few reddish lamellae.

F OR-7 115 - 120 Single continuous stratum of 5 YR 5/4 sand, no increase in clay
or field-detectable change in sand texture, sampled
separately.

F OR-7 150 - 200 Few possible lamellae.

FOR-8 none

FOR-9 90-120 Lamellae 7.5 YR 4/4, 2 mm thick, 10 cm apart. Some thin black
strata.

FOR-9 270 - 290 Silty stratum, sampled separately.

SAV-l 180 Few fine lamellae around a gravel lens.

SAV-2 260 Few very faint lamellae.

SAV-3 90 - 180 Continuous and level bands 7.5 YR 5/6, < 1 mm thick, 2 cm
apart, very little clay.

SAV-4 120 - 180 Very thin bands, some orange, some black, no clay increase.
Two thicker orange strata 2 mm thick, 12 cm apart.

SAV-S 120 Single 4 mm thick band of orangish sand.

SAV—6 110 - 230 Very thin reddish and black bands.

SAV-7 170 - 250 Possible few fine lamellae.

3V8 none

 

95

Table 13. Solum weighted particle size distributions for sampled pedons.

 

 

 

 

Sand Silt Clay VCS CS MS FS VFS
2.0-0.05 0.05-0.002 <0.002 2.0-1.0 1.0-0.5 0.5-0.25 0.25-0.1 0.1-0.05
Pedon mm mm mm mm mm mm mm mm
------ % of fine earth------ ---------------% of sand fraction---------------
FOR-1 97.6 1.4 1.0 2.9 14.0 56.3 25.4 1.3
FOR—2 97.6 1.5 0.9 1.8 10.3 62.3 24.6 0.9
FOR-3 97.8 1.3 0.9 0.7 5.2 55.5 36.3 2.3
FOR-4 97.2 1.5 1.3 1.4 9.8 62.4 25.6 0.8
FOR-5 96.8 1.8 1.4 1.5 12.2 66.0 19.6 0.7
FOR-6 95.5 3.2 1.3 1.4 10.4 55.0 29.2 4.0
FOR-7 97.8 1.3 0.9 2.0 14.0 61.3 22.1 0.7
FOR-8 96.4 2.1 1.5 7.7 24.0 52.0 15.2 1.1
FOR-9 96.7 2.0 1.3 0.6 5.7 50.9 40.5 2.3
Mean1 97.0 1.8 1.2 2.2 11.7 58.0 26.5 1.6
SAV—l 97.6 1.5 0.9 3.1 16.6 54.9 24.2 1.1
SAV—2 98.3 1.0 0.7 3.7 16.6 58.8 20.2 0.7
SAV-3 96.9 2.0 1.1 1.1 5.8 57.6 33.1 2.3
SAV-4 97.4 1.7 0.9 1.3 9.6 50.6 36.6 1.9
SAV-5 97.0 2.0 1.0 3.4 8.8 52.0 i 33.8 2.0
SAV-6 97.7 1.3 1.0 0.5 3.8 59.1 35.7 0.9
SAV-7 97.6 1.5 0.9 2.0 17.3 57.1 22.7 0.9
SAV—8 97.0 1.7 1.3 1.7 11.6 60.3 25.3 1.1
Mean] 97.5 1.6 1.0 2.1 11.3 56.3 29.0 1.4

 

IMeans are not significantly different at p < 0.05 (Kolmogorov-Smirnov) for all columns.

96

due to weathering and soil development processes. Means for all solum weighted particle
size fractions are not significantly different between forested and stump prairie pedons
(Table 13). Mean clay content is slightly higher in forested pedons than in stump prairie
pedons, but the difference is not significant at p < 0.05 (Kolmogorov-Smimov). This
slight difference is due to a small increase in clay in the B horizon, possibly illuvial in
nature, as discussed below (p. 112). Therefore, while the parent materials of forested
soils and stump prairie soils show no textural differences, soil development processes
may have caused a slight elevation of what was initially a very low clay content in the
forested soils. The increased clay content, though small, may increase the water holding
capacity and CEC of the soil.

W For most horizons in both forest and stump prairie
soils, the sand fraction distribution is maximal in the medium sand size class, but in a few
cases it falls instead in the fine sand size class (Table 22 in Appendix A). Although the
dominant sand fraction, the percent medium sand can vary erratically within a single
pedon (Figure 28), as can the ratio of coarse to fine sands (Figure 29). No statistically
significant difference between forested and stump prairie soils exists for any sand size
fraction (Table 13), nor for the ratios between the coarser and finer sand fractions, either
in the C horizon or weighted for the solum as a whole (Table 14). Within each group,
however, there is much variability in these ratios (Table 14).

The slight variations in sand fraction distributions within and between pedons are
to be expected in an outwash environment where depositional energy can vary widely
over time, leading to numerous variations in grain size both horizontally and vertically
(Flint 1971, p. 186). In that sense these pedons can be seen as typical of soils developed
in outwash sands (Schaetzl 1992). The lack of significant differences between the two
groups suggests that differences in the sand size fractions are unlikely to have affected

forest regeneration following nineteenth century logging.

 

97

Percent MS

 

 

 

 

 

 

 

 

 

A Percent MS B
0 20 40 60A 80 100 0 20 40 60 80 100
o ——--9»——— . A T 0 +—————--— .4.
E E
BS1 BS1
50 - 50 + 882
B52
BC
BC
100 - 100
A A C
E E .
2. . 3 .
j 5 150 e 1 5 150 +
i 9 0 ~ 8 T
; 0 . i 0 ,
I 200 7+ 200 +
250 + 250 ~
3 Deep C
300 + SAV-Z 300 4 SAV-8
C Percent Ms 1 D Percent MS
0 20 40 60 80 100 = O 20 40 60 80 100
0 +1 0 ~— —— +
E E
i Bhs
' 50 j 882 ' j 50 + 852
! so 1
BC ‘ L
. 100 ~ 2 i 100 ~ C 1
i a. j g 1.
1 ~5— C 3% .5. . a
i .1: 150 + . * 150 3 : ;
1 a 1 . :1 i . .
0 | I o e
’- O . . f I O . ’
200 .; 6 Deep 0 . ‘ 200 4}
- 1
i j ‘ DeepC
250 4 250 ;
I ;
300 4 FOR '2 300 i FOR -5

 

Figure 28. Percent medium sand by depth for four representative pedons. Error bars
signify high and low values in horizons where ortstein was sampled separately.

A CSIFSratio
0.0 A 2.0 4.0 6.0

B CSIFS ratio
0.0 2.0 4.0 6.0

50 -+.-

   

 

 

 

 

100 - 100 ,
- c
‘ E». . .5.
‘ .c 150 - ' s 150 +
t '5. 1 '5.
l 8 l 3
200 + 1 200
250 i ‘ 250 T
, 1 1 '
? 1’ Deep C
300 a SAV-Z % 300 .I. SAV—8
l
C CSIFS ratio i D CSIFS ratio
0.0 2.0 4.0 6.0 i 0.0 2.0 4.0 6.0
S ; .
i 351 1 ; i 851
50 . 352 1 3 50 I 382
1 BC . BC
100 I: 1001 , C
. ‘e‘ C i ‘2‘ i
. 9. l .9. O .
, 150 4 5 .1: 15 «.—
o 7 ' l n l
1 . 1 E
200 4‘ Deep C : , 200 *j .
j g ; . 4 Deep C
250 i. j I 250 4
300 4 FOR '2 ; ; 300 L FOR-5

[ I
1

Figure 29. Ratio of the coarse sand fraction t6 fie fine sand fraction by depth for four *

representative pedons. Error bars signify high and low values in horizons where
ortstein was sampled separately.

99

Table 14. Ratios between coarse and fine sand fractions.

 

 

 

 

 

C horizon Solum weighted

Coarser/ Coarser/

Pedon CS/Fsl Finer2 CS/FS' Finerz
FOR-1 0.86 3.96 0.55 2.75
FOR-2 1.52 8.37 0.42 2.92
FOR-3 0.19 3.18 0.14 1.59
FOR-4 0.22 1.49 0.38 2.79
FOR-5 0.92 7.45 0.62 3.93
FOR-6 0.60 2.35 0.36 2.01
FOR-7 2.10 9.01 0.63 3.39
FOR-8 1.08 4.69 1.58 5.14
FOR-9 0.50 2.77 0.14 1.34
Mean3 0.89 4.81 0.54 2.87
SAV-l 0.21 1.57 0.69 2.95
SAV—2 0.38 6.34 0.82 3.78
SAV—3 0.01 0.62 0.18 1.82
SAV-4 1.96 12.24 0.26 1.60
SAV-S 0.18 2.05 0.26 1.80
SAV—6 0.03 1.30 0.1 1 1.73
SAV-7 0.55 3.63 0.76 3.24
SAV—8 4.42 10.00 0.46 2.79
Mean3 0.97 4.72 0.44 2.46

 

lRatio of coarse sand to fine sand fiactions.
2Ratio of (very coarse sand + coarse sand + medium sand) to (fine sand + very fine sand).
Means are not significantly different at p < 0.05 (Kolmogorov-Smirnov) for all columns.

100

Wm

B horizon properties affected by soil development processes show some
statistically significant differences between the forest and stump prairie (Table 15). This
section will discuss how the soil characteristics affected by developmental processes
relate to the vegetation patterns in the Kingston Plains. The next section (p. 112) will
consider the implications of these data to an understanding of soil development processes.

Differences between forest and stump prairie pedons in weighted B horizon
means for CC, ODOE, Feo, Fep, and Ald are statistically significant; means are higher in
forest pedons (Table 15). Higher weighted B horizon means of 0C and some forms of Fe
and Al in forest soils suggest stronger spodic horizon development in these pedons due to
illuvial accumulations of OC, Fe, and Al. Solum weighted means for ODOE, F e0, Fep,
Ald, and Alp are also significantly different between forest and stump prairie (Table 16).

Pedons in the forested areas have more ortstein than pedons in the stump prairie
(mean le ortstein content of forest pedons is 39.4%; of stump prairie pedons, 22.5%).
Forest and stump prairie pedons have no statistically significant difference in OC and
extractable Fe and Al contents of ortstein subsamples of B51 horizons (Table 17).
Likewise, le matrix subsamples are not significantly different in OC, Fe, and Al
contents between forest and stump prairie pedons (Table 17). Mean OC, ODOE, and Fep
contents, however, are higher for all BS] ortstein subsamples than for B51 matrix
subsamples; Sio is higher in matrix than in the ortstein subsamples (Table 18). Although
not statistically significant, means for other forms of A1 and Fe are also higher in ortstein
than the matrix subsamples, except for A10. Therefore, it appears that the higher weighted
B horizon means for OC, Fe, and A1 in the forested pedons over the stump prairie pedons
may be primarily associated with higher ortstein content in the former.

Like OC, ODOE, and F ep, mean clay content is significantly higher in le

Ortstein subsamples than in matrix subsamples for all pedons (Table 18), but there is no

101

Table 15. B horizon weighted means for selected soil properties.

 

 

Pedon Silt Clay 21111120 ODOE 0C Fed Feo Fep Ald Alo Alp
...... 919------ 2.1121
FOR-1 1.3 1.3 4.9 0.131 3.4 1.1 0.7 0.4 1.0 1.7 1.0
FOR-2 1.4 1.3 5.0 0.122 3.3 1.5 0.9 0.5 1.3 2.0 1.1
FOR-3 1.2 1.3 4.7 0.228 5.5 1.3 0.8 0.6 1.2 1.7 1.1
FOR-4 1.7 1.9 4.6 0.276 6.9 2.0 1.4 0.7 1.6 2.8 1.3
FOR-5 1.5 1.7 4.8 0.191 6.1 1.6 1.1 0.7 1.5 2.1 1.8
FOR-6 3.0 1.3 5.1 0.133 4.5 1.8 1.1 0.4 1.3 2.5 1.3
FOR-7 1.0 1.4 5.2 0.197 5.4 1.5 1.0 0.6 1.3 2.7 1.5
FOR-8 2.0 2.0 5.1 0.252 7.3 1.9 1.2 0.8 1.8 2.9 2.1
FOR-9 2.2 1.8 4.9 0.239 6.7 1.7 1.0 0.7 1.5 2.2 1.5
Mean 1.7 1.6" 4.9. 0.196. 5.5" 1.6 1.0” 0.6" 1.4. 2.3 1.4
SAV—l 1.2 0.9 5.4 0.056 1.6 0.9 0.5 0.3 0.8 1.6 0.9
SAV-2 0.7 0.7 5.5 0.059 1.5 0.8 0.5 0.2 0.6 1.6 0.8
SAV-3 2.0 1.2 5.2 0.107 4.2 1.6 0.8 0.4 1.4 2.3 1.5
SAV-4 1.6 1.1 4.6 0.093 3.2 1.5 0.8 0.3 1.1 2.7 1.0
SAV-S 1.8 1.1 5.3 0.081 2.7 1.8 0.8 0.4 1.2 2.5 1.1
SAV-6 1.2 1.2 5.1 0.195 4.5 1.5 0.9 0.5 1.3 2.4 1.1
SAV—7 1.2 1.2 5.2 0.153 4.2 1.3 0.8 0.4 1.2 2.7 1.0
SAV—8 1.3 1.5 6.0 0.126 3.3 1.4 0.8 0.3 1.2 2.4 1.0
Mean 1.3 1.1” 5.3. 0.109. 3.1” 1.3 0.7” 0.3" 1.1. 2.3 1.1

 

 

 

 

 

:Differences in column means statistically significant at p < 0.05 (Kolmogorov-

Smimov).

"Differences in column means statistically significant at p < 0.01 (Kolmogorov-

Smimov). All other, unmarked columns, are not statistically significant.

102

Table 16. Solum weighted means for selected soil properties.

 

 

 

 

 

 

Pedon Silt Clay 211311—1120 ODOE OrganicC Fed Feo Fep Ald Al0 AlL
------ 2. 21:91
FOR-1 1.4 1.0 4.9 0.081 2.5 0.9 0.4 0.3 0.7 1.2 0.7
FOR-2 1.5 0.9 5.0 0.082 2.9 1.1 0.6 0.3 0.9 1.4 0.8
FOR-3 1.3 0.9 4.9 0.140 4.0 0.9 0.5 0.4 0.7 1.1 0.8
FOR-4 1.5 1.3 4.9 0.173 4.7 1.3 0.9 0.5 1.1 1.9 0.9
FOR-5 1.8 1.4 4.9 0.123 5.2 1.1 0.7 0.5 1.0 1.4 1.2
FOR-6 3.2 1.3 5.1 0.087 5.9 1.3 0.8 0.3 0.8 1.6 0.9
FOR-7 1.3 0.9 5.3 0.106 3.7 0.9 0.6 0.4 0.8 1.5 0.9
FOR-8 2.1 1.5 5.1 0.166 5.5 1.4 0.8 0.5 1.2 2.0 1.5
FOR-9 2.0 1.3 5.0 0.148 4.7 1.2 0.6 0.5 1.0 1.5 1.0
Mean 1.8 1.2 5.0” 0.123“ 4.3 1.1 0.7. 0.4" 0.9. 1.5 1.0.
SAV-I 1.5 0.9 5.4 0.043 3.9 0.7 0.4 0.2 0.6 1.2 0.7
SAV—Z 1.0 0.7 5.5 0.046 2.4 0.7 0.4 0.2 0.5 1.3 0.7
SAV—3 2.0 1.1 5.2 0.067 5.0 1.0 0.5 0.3 0.9 1.5 1.0
SAV-4 1.7 0.9 4.9 0.058 3.8 1.1 0.5 0.2 0.7 1.7 0.7
SAV-5 2.0 1.0 5.2 0.047 4.5 1.1 0.5 0.2 0.7 1.5 0.7
SAV-6 1.3 1.0 5.1 0.129 4.3 1.0 0.6 0.4 0.9 1.7 0.9
SAV—7 1.5 0.9 5.3 0.084 4.5 0.8 0.4 0.2 0.7 1.5 0.7
SAV-8 1.7 1.3 5.7 0.062 4.8 0.8 0.4 0.2 0.6 1.2 0.6
Mean 1.6 1.0 5.3” 0.067” 4.1 0.9 0.5. 0.2“ 0.7. 1.5 0.7.

 

*Differences in column means statistically significant at p < 0.05 (Kolmogorov-
Smimov).

"Differences in column means statistically significant at p < 0.01 (Kolmogorov-
Smimov). All other, unmarked columns, are not statistically significant.

103

Table 17. Means of BS] ortstein and BS] matrix subsamples for selected soil properties
compared for forest and stump prairie soils.

 

 

 

Ortstein Matrix
Soil property Forest Stump Prairie Forest Stump Prairie
pH (2:1 H20) 47* 52* 4.7* 52*
Clay % 2.7 2.6 2.0 1.7
Silt % 2.3 2.3 2.5 2.6
oc g kg" 9.6 9.1 6.1 4.4
ODOE 0.425 0.415 0179* 0128*
Fed g kg" 2.3 2.7 2.2 2.1
Feo g kg" 1.6 1.7 1.4 1.1
Fep g kg" 1.1 1.0 0.6 0.4
A1,, g kg" 2.1 2.7 2.0 1.7
14.10 g kg" 2.8 3.6 3.9 3.6
A1,, g kg" 1.9 1.4 1.8 1.5
81, g kg" 1.2 1.6 2.4 2.4
Pea/Fed 0.70 0.63 ‘ 067* 054*
(Fee-FeP)/Fep 0.55 0.72 1.78 1.62
4.10/41d 1.33 1.33 1.97 2.27
(Ale-Alp)/Alp 0.49 0.57 1.14 1.46
Ala-Alp 0.87 1.24 2.1 1 2.13
(AID-Alp)/Sio 0.85 0.83 0.91 0.89
3,011, 0.29 0.22 0.14 0.14

 

*Means significantly different at p < 0.05 (Kolmogorov-Smirnov).

104

Table 18. Selected mean soil properties of B31 ortstein and 851 matrix subsamples for
all the sampled pedons.

 

 

Soil Property Ortstein Matrix
pH (2:1 H20) 4.9 4.9
Clay % 2.6" 1.8**
Silt % 2.3 2.5
oc g kg" 94* 53*
ODOE 0420* 0.155
1=ed g kg" 2.5 2.2
Fe0 g kg"1 1.7 1.3
Fep g kg" 1.0* 0.5
A1,, g kg'1 2.4 1.9
Al0 g kg" 3.2 3.7
A1,, g kg" 1.7 2.1
Sio g kg'1 1.4* 2.4
Fee/Fed 0.67* 0.59
(F eo-Fep)/F ep 0.63 1 .70* *
AID/A1,, 1.33 2.10"
(Al(,-All,,)/Alp 0.52 l .29* *
AID-Alp 1.04 2.12"
(A1(,-Alp)/Sio 0.84 0.90
icy/Alp 0.26" 0.14

 

*Means significantly different at p < 0.05 (Kolmogorov-Smimov).
"Means significantly different at p < 0.01 (Kolmogorov-Smimov).

105

statistically significant difference between forest and stump prairie in ES] ortstein and
matrix subsamples considered separately (Table 17). Silt content is essentially the same
for both ortstein and matrix samples, and pH is significantly different between stump
prairie and forest pedons, but not between ortstein and matrix subsamples (Tables 17 and
18). Thus, clay content, like OC, Fe, and Al contents, appears to be associated closely
with the presence of ortstein, such that the differences in B horizon weighted mean clay
content between forested and stump prairie pedons is primarily dependent on the
differences in estimated ortstein content in the two soil groups. In contrast, differences in
pH between the two groups do not depend on ortstein content, but are primarily
associated with differences in current landcover type.

Illuvial accumulations of OC and free Fe in the spodic horizons of sandy soils are
associated with increased water availability in the soil and with increased forest growth
(Shetron 1974). In Michigan, strength of spodic horizon development has also been
associated with forest type, with jack pine forests commonly found on weakly developed
Udipsamments, and white pine and hardwood forests on the more strongly developed
Spodosols (Mokma and Vance 1989). There are, therefore, two possible explanations for
the association of greater illuvial accumulations of OC, Fe, and A1 with currently forested
areas:

(1) The greater water retention of more strongly developed soils was more

favorable to tree regeneration and forest recovery following logging and,

therefore, the patterns of soil development directly influenced patterns of

forest regeneration.

(2) Patterns of spodic horizon development are related to differences in

pre-logging forest. Areas that today are forested were richer in hardwoods

than were current stump prairie areas (p. 62). The differences in pre-

logging forest type may have influenced forest recovery patterns

independently of spodic horizon development.

106

The present data do not allow me to choose between these two possible explanations;

most likely both have interacted and played a part.

107

Transect:

In order to complement information derived from the 17 sampled pedons,
transects spanning the forest/stump prairie boundary were completed to provide further
data on both ortstein content and soil B horizon color of the forest and stump prairie soils.
Because ortstein content appears to be important to the relationship between soil
characteristics and forest recovery, transects were thought to provide an additional means
to evaluate whether the differences in ortstein content observed in the sampled pedons are
representative, and whether the ortstein content changes near the forest/stump prairie
boundary.

For all transects, ortstein content is greater on the forested side of the boundary
than on the stump prairie side (p < 0.001 , Kolmogorov-Smimov; Figure 30). Similarly,
the darkest, reddest B horizon color has lower (redder) Munsell hue and lower (darker)
value and lower chroma on the forested sides of the transects (Kolmogorov-Smimov p <
0.001; Figures 31 - 33). Increasing ortstein content in sandy soils is associated with
increasing spodic horizon development (Wang et al. 1978; Barrett and Schaetzl 1992,
1993). In Spodosols, redder hue and darker value are indicative of stronger soil
development (Schaetzl and Mokma 1988). Hue becomes redder and value and chroma
decrease with increasing amorphous organo-metallic complexes (Mokma 1993), and with
increasing soil age (Barrett and Schaetzl 1993) in Michigan Spodosols. In Ontario,
redder hues and lower values are correlated with Fe accumulation, and lower chromas
with OC accumulation (Evans and Cameron 1985).

The contrasts in ortstein content and color on either side of the forest/stump
prairie boundary are most noticeable for transects 4 and 6, both of which are located
along the southern border of the stump prairie area (Figure 30). Transect 4 is located in a
region where the boundary has remained relatively stable between 1939 and 1986 (see p.

49), but the boundary near transect 6 has shifted, such that the current vegetation

Forest Stump Prairie

    

Transect 1
% Ortstein

 

 

Transect 2

 

 

 

 

 

Transect 3
‘16 Ortstein

 

 

 

 

Transect 4
'16 Ortstein

 

 

Transect 5
96 Ortstein

 

 

 

100+
75+

Transect 6
% Ortstein

 

180 150120 90 60 3O " 30 60 90 120 150180

Distance (m) from boundary 1

J

Figure 30. Percent of sample points in which ortstein was detected for each quadrat on
forest/stump prairie transects.

109

5 Forest Stump Prairie

I-lllllllll
1111111

Transect 1
Hue
0)

0|

4

 

      

7

Transect 2
Hue

 

 

 

Transect 3
Hue

 

Transect 4

 

 

 

Transect 5

Transect 8
Hue

 
 
   

       

90

 

"1' fl

180 150 .120

2‘
N
U
C
__=_
O
a:

60 30 30 60 90 120150 180 1
Distance (m) from boundary 1

 

Figure 31. Arithmetic mean of recorded Munsell hue for the sample points in each
quadrat on forest/stump prairie transects.

Forest Stump Prairie

   

 

 

Transect 2
Value
(a)
'a

 

 

     

 

 

 

 

 

 

 

Transect 4

 

 

 

 

 

 

 

 

    
 

 
      

IO

*5 .
8 :
c 1
5 1
'— |
CD

15

Q

m

5 32
3 m'. User 3

O

180 150120 90 6O 30 ‘m 30 60 90 120 150180
Distance (m) from boundary

 

Figure 32. Arithmetic mean of recorded Munsell value for the sample points in each
quadrat on forest/ stump prairie transects.

Forest Stump Prairie

Jhl
Jr] I .l -I-

  
 

 

 

 

 

 

I ' l
" !
V 'v
‘ .,
., 1
4.
. . .
.._
‘ _5__ _

 

 

 

E" 1'

 

 

 

Transect 5
Chroma

   

 

 

   

. 180 150 120 90 60 30 .3 ‘ 30 6O '90 120150 180
1 Distance (m) from boundary

Figure 33. Arithmetic mean of recorded Munsell chroma for the sample points in each
quadrat on forest! stump prairie transects.

112

boundary does not coincide precisely with the boundary as it was previously established.
Nevertheless, based on the aerial photographic analysis of boundary stability (p. 44), it is
probable that forest/stump prairie boundaries for all transects have not moved enough
since the boundaries were established to be discerned by the 30 m spacing between
transect samples.

Both soil color and ortstein content along these transects provide evidence that the
soils in forested regions are currently more strongly developed than the soils of the stump
prairie. Based on the transect data, it appears that the pedologic boundary between better
developed and less developed soils approximately coincides spatially with the present

biotic boundary.

Soil development processes

In this section, soil development processes in the study area are examined. First,
soil morphological and chemical data will be discussed relative to soil forming processes.
Then the issue of whether soil development processes are different between forested and
stump prairie areas will be examined in light of both soil properties (solid phase data) and
information on soil solution composition (active phase data).

W

Clay depth plots demonstrate that, in both stump prairie and forest soils, a
subsurface maximum in clay content is usually reached in the upper B horizon. Stump
prairie pedons also often have a surface maximum in the A horizon (Figure 34). The clay
content maximum in the upper B horizon, coupled with the association of ortstein
subsamples from the B31 horizon with higher clay content (Table 18), suggests that the
clay is being translocated into the B horizon from the surface horizons. Slight B horizon
increases in clay, apparently illuvial in origin, have been observed in Spodosols and

Spodosol-like soils in many areas (Brydon 1965; Wang and Rees 1980; Stanley and

113

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Percent Clay 3 Percent Clay
A 0.0 2.0 4.0 6.0 0.0 2.0 4.0 6.0
0 ‘ IL 5 9 i 0 i g 1
. E A e/. A
tl—l '\ E 1
9 s
50 4, B 2 881 50 "' /'°‘{52.
* s
‘ ° BC
100 so 100 1
5. .5. :
.: 150 * 5150 -~:
3';- . c 8' i
o ' o :
200 -- 200
250 -- 250 «E
éDeep C
300 .- SAV - 2 300 -1 SAV - 8
C Percent Clay D Percent Clay
0.0 2.0 4.0 8.0 0.0 2.0 4.0 6.0
O + E 1 4. O 1 E + :
.7 Bhs 4>.: 851
st
50 -- . 352 5° " ) asz
. BC
./ BC 1
100 -- 100 c
E. E C .5. 5
1:150 ~-- 3150 ...
g. .' E. t
200 —2‘ Deep C 200 +1
1 Deep C
250 -- 250 -.
300 -~ FOR '2 300 .. FOR -5

 

Figure 34. Percent clay by depth for four representative pedons. Error bars signify high

and low values in horizons where ortstein was sampled separately.

 

114

Ciolkosz 1981). F ranzmeier (1962), Nomberg (1977), and Barrett and Schaetzl (1992)
all reported slight clay increases with increasing soil age in B horizons of podzol
chronosequences.

As is typical for many sandy soils undergoing podzolization, soils in both forest
and stump prairie have a silt content maximum in the surface horizon (Figure 35). A
similar silt maximum has been reported for Spodosol soils in many locations, and has
been shown to increase with the age of the surface on which the soil is developing
(Franzmeier 1962; Jauhiainen 1973; Nomberg 1977; Barrett and Schaetzl 1992).
Franzmeier (1962) and Narnberg (1977; 1980) concluded that silt-rich surface horizons
are the result of in situ physical weathering of sand-sized particles. An eolian origin for
the silt cannot be ruled out in the present study, but surface weathering of sand-sized
particles is the most likely explanation.

For both forest and stump prairie pedons, pH increases with depth (Figure 36).
Higher acidity in the surface horizons, especially in the O horizons of forest pedons, is
probably associated with the presence of organic acids. Acidity of the surface mineral
horizons can also be associated with an increase in mineral weathering rates in these

horizons.

115

 

Percent Silt Percent Silt
A 0.0 2.0 4.0 6.0 B

0.0 2.0 4.0 6.0

 

 

 

Depth (cm)

 

 

 

 

 

 

 

 

250 «~
Deep C
300 ~- SAV - 2 300 SAV - 8
' Percent Silt D Percent Sllt
C 0.0 2.0 4.0 6.0 0.0 2.0 4.0 6.0
0 i l E 1‘ O i i E i
Bhs 851
50 - 50 -.
852
BC
100 100 , C
‘e‘ ‘5‘ :
3 9., 1
.c 150 150 ~-
§ § :
a 3
200 Deep C 200 “i
1tDeep C
250 ~- 250 «-
300 1 FOR - 2 300 -- FOR - 5

 

 

 

 

 

 

 

Figure 35. Percent silt by depth for four representative pedons. Error bars signify high
and low values in horizons where ortstein was sampled separately.

116

 

 

 

 

 

 

 

 

 

 

 

 

 

 

pH in water B pH in water
A 3.0 4.0 5.0 6.0 7.0 3.0 4.0 5.0 6.0 7.0
0 $A9‘3i : i O iA,\. E i 4
cl 851 \M 851
50 w \ 50 ~- .o' 852
. 852 1
° BC
100 ~- ° BC 100 11 \
. A 9 C
E E .'
2. 2, .
5150 ~- 5150 ._ ,-
Q o c a
3 ' _ o :
200 -- 200 -- f
250 q- 250 «~ 5
2' Deep C
300 -- SAV - 2 300 -L SAV . 3
C pH in water D pH in water
3.0 4.0 5.0 6.0 7.0 3.0 4.0 5.0 6.0 7.0
O .4 * l l : .. : : i
Oa'\t E O O" E
°\ Bhs \.
,. . 851
1" 851 \
so -- \ 852 5° " rugsz
\ so i BC
100 -- 1 100 -- z c
5150 ._ f 5150 -- 3
g. . a .
a 3
200 -- § Deep C 200 ~r- '.
. Deep C
250 .. 250 .-
300 -- FOR - 2 300 “ FOR - 5

 

Figure 36. pH by depth for four representative pedons. Error bars signify high and low
values in horizons where ortstein was sampled separately.

 

 

117

W

In this section, I discuss evidence of podzolization processes in the solid phase of
the sampled pedons, and whether any differences exist between pedons under forest
vegetation and those in the stump prairie. The distribution within a pedon of forms of Fe,
Al, and 0C are indicators of podzolization. Solid phase soil properties, however, may
provide evidence of processes that are no longer active in the soil. In the next section I
will discuss evidence for whether podzolization processes are active in the study area.

In the sampled pedons, a large 0C surface maximum is present, and also a
smaller, subsurface maximum in the upper B horizon (Figure 37). The presence of some
0C in the B horizon coupled with relatively low le horizon 0C content masks the B
horizon maximum in some stump prairie pedons (Figure 37A). The fulvic acid portion of
organic matter has been theorized to be important in the podzolization process (De
Coninck 1980). ODOE, which is related to fulvic acid content (Daly 1982), shows a
large subsurface maximum in the uppermost B subhorizon of most pedons, with very low
values in the E horizons of both stump prairie and forested pedons (Figure 38). Both 0C
and ODOE are higher in ortstein than in non-ortstein subsamples from the same horizon
(Table 19).

All three forms of extractable Fe and Al usually have subsurface maxima in the B
horizon (Figures 39 and 40). The maximum Fe content occurs in the uppermost B
subhorizon, but the maximum content of extractable Al is in the B51 horizon for those
pedons with Bhs horizons (Table 23, Appendix A), lower in the profile than the Fe
maximum. Al maxima that occur deeper than the corresponding Fe maxima have been
reported of podzols in the USA (Franzmeier and Whiteside 1963; Barrett and Schaetzl
1992), Japan (Mizota 1982), Finland (Koutaniemi et al. 1988), and Quebec (DeKimpe
and Martel 1976; Kodama and Wang 1989). This phenomenon may occur because Al is
more mobile than Fe, and thus initially deposited at greater depth (Mizota 1982), or

118

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

c kg"
A 0.0 5.0 13.0 15.0 20.0 B
0 i l L a 0
/‘ E A
I--——I 351
50 1* 50 .
1 852

100 11 BC 100 ~
‘5‘ E
a 2.
5150 .. 5150
E 1. :-
a C a

200 -- 200

250 -- 250

300 -- 5Av - 2 300

oc kg" oc kg"
C 0.0 5.0 1 .0 15.0 20.0 D 0.0 5.0 1 .0 15 0 20.0
0 E i l ‘ fi’ 0 Ei % i 1
° A 2 fl
-. 2 Bhs ‘4 BS1
851
50 q. 50 ._ ./
i 852 1' 352
2 BC

100 -i BC 100 -~ \, c

150 «J 5150 -- -'
g. : a :
o 3

200 5 Deep C 200 ~-.'

°' Deep C
250 .. 250 «-
300 -- FOR -2 300 .. FOR -5

 

Figure 37. 0C content by depth for four representative pedons. Error bars signify high
and low values in horizons where ortstein was sampled separately.

 

119

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A ODOE B ODOE

0.0 0.2 0.4 0.6 0.0 0.2 0.4 0.6

0 ?\E 1 1 4. 0 9 E 1 1 1
'—° . 851 °——'1 851
50 T 50 -- 7° 852
IF B52 . BC

100 «'1 BC 100 \
A A 9 C
5 s ;
5150 - '5'150 1- :
Q Q
8 9 C 8 ..

20° " 200 -- :

250 ~- 250 17'

.1 Deep C

300 -- SAV - 2 300 SAV - s

c ODOE D ODOE
0.0 0.2 0.4 0.6 0.0 0.2 0.4 0.6
0 3 1 3h 1 O . E 1 1 41
5?; 1 s .—1 851
5
50 -v- / 50 "" 7. 852
° BSZ
/ 33..

100 I BC 100 c
E : C E I
5150 «j 5150 «1
° : 8 :

200 Deep C 200

§ Deep C
250 ._ 25o ._
300 -~ FOR - 2 300 «L FOR - 5

 

Figure 38. ODOE values by depth for four representative pedons. Error bars signify high

and low values in horizons where ortstein was sampled separately.

 

120

Table 19. Summary of selected mean soil property relationships with regard to current
landcover type and soil ortstein status.

 

 

Mean is higher in:
Soil property Forest or Stump Prairie? Orstein or Matrix?
pH 2:1 H20 Stump Prairie (B, S)‘ --
Clay % Forest (B, S) OrtsteinM
Silt % Forest Matrix
0C g kg'1 Forest (B) Ortstein“
ODOE Forest (B, S) Ortstein"
Fed g kg'1 Forest Ortstein
Feo g kg'1 Forest (B, S) Ortstein
Fep g kg'I Forest (B, S) Ortstein
Ald g kg"1 Forest (B, S) Ortstein
Al0 g kg'1 Forest Matrix
Alp g kg‘1 Forest (B, S) Matrix
Sio g kg'1 Stump Prairie Matrix“
Fee/Fed Forest Ortstein“
(Fe,,-Fep)/Fep Stump Prairie (B) Matrix"
Ala/Ala Stump Prairie (B, S) Matrix"
(Al(,-Alp)/Alp Stump Prairie (B, S) Matrix"
Ale-Alp -- Matrix"
(Al(,-Alp)/Sio Stump Prairie Matrix
Pep/Alp Forest (B, S) Ortstein"

 

"‘B” indicates a statistically significant difference (p < 0.05, Kolmogorov-Smimov) in
weighted B horizon values; “S” indicates a statistically significant difference (p <

0.05, Kolmogorov-Smirnov) in weighted solum values.
* p < 0.05, Kolmogorov-Smimov
** p < 0.01, Kolmogorov,Smirnov

 

 

 

0 kg"

2.0

4.0 5.0

3.0

 

 

.1
9|!!!

 

1 .0 2.0 3.0 4.0 5.0

0.0

 

 

“-...-.“ Pep
— ~o- - Fed

--—-D-- Feo

 

 

 

 

 

 

121

.1
9 k9

 

2.0 3.0 4.0 5.0

1.0

0.0

   

‘0'.

 

0 ko“

.bamlglillwng

b
1 q q

mWMm m m

30011

 

 

4.0 5.0

3.0

2.0

 

 

 

 

Figure 39. Extractable Fe content by depth for four representative pedons.

122

 

0 to"
2.0

5.0

4.0

3.0

1.0

0.0

 

 

 

 

 

 

 

Wit-- Alp

... .... . Nd
---a-- Ale

 

 

 

 

 

 

 

5.0

4.0

9 Re"
3.0

2.0

1.0

 

 

 

 

0.0

 

 

 

0.
5
0.
‘1
w
..w.
0 0
Z
0.
_..1rmww...u..w..1u...unuEwa1 m
m l m
1mm- m m m o
0.
5
D
4
m 1;. w;
49 x x
.u a ...
N 1 @101 /
rah? .,
0 1 m 1......mu/z
.. p/ 1 ‘ 0.]..u1m1uumwrr.
.wu. Hummwwwm a n L. M m 1 1
m m1. w o m m m1.
.. m .1. .. m .1. m c m

 

3001

 

Figure 40. Extractable Al content by depth for four representative pedons.

123

because the Al is preferentially remobilized after deposition and translocated still deeper
(DeKimpe and Martel 1976; Farmer 1984). The depth distribution of Sio is similar to that
for Al (Figure 41).

Comparisons of extractable forms of Fe and Al can provide insight into the nature
of the Fe and Al in soils. Fep is considered to represent organically-bound, amorphous
forms of Fe; Feo represents amorphous Fe, both organic and inorganic; and Fed is “free”
Fe, both amorphous and crystalline (McKeague and Day 1966; McKeague 1978).
Interpretation of extractable Al is analogous, but the exact nature of Ald is not well
understood, and A10 content generally exceeds Ald in the same horizon (McKeague and
Day 1966; McKeague et al. 1971; Parfitt and Childs 1988). Farmer et al. (1983) have
shown that, while acid armnonium oxalate is an effective extrath for poorly ordered
phases of Fe and Al, only an ill-defined fraction of any allophane-imogolite complex in
the soil is extracted by DCB. Oxalate-soluble Si (Sic) is considered to represent
amorphous Si forms such as are present in allophanes (Farmer et al. 1983; Parfitt and
Kimble 1989).

The relative crystallinity of free iron oxides can be measured by the “activity
ratio” or Fee/Fed (McKeague and Day 1966; Blume and Schwertrnann 1969). Within a
given profile, a lower Feo/Fed ratio indicates a larger proportion of crystalline Fe oxide
compounds and, therefore, lower Fe mobility in the soil (Singer et al. 1978). The depth
distribution of F eo/F ed in the sampled pedons usually shows a maximum in the upper B
horizon (Figure 42), suggesting that the extractable Fe is, or recently has been, most
mobile in the upper B horizon. In the E and C horizons, extractable Fe is in a relatively
crystalline form. This finding differs from that of Singer et al. (1978), who reported that
the minimum F eo/F ed value was in the upper B horizon of a Spodosol in the Cascades,
possibly due to Fe recrystallization and retention in the B horizon. Within the B51

horizon, a statistically greater Fee/Fed ratio occurs in ortstein over non-ortstein

124

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

51 k " 51 g k "
A 0.0 1.0 2.69 :30 4.0 5.0 B 0.0 1.0 2.6 39.0 4.0 15.0
0 K 1 1 1 1 0 f5 1 1 1 1 1
e-I BS1 >._‘ BS1
50 -- / 50 ~~ / 352
/ 852 0 BC
100 ""80 100 --/
.5. E. :
5150 -- 5150 --:
a. 06 g :
3 a :
200 1- 200 --§
250 ~- 250 «I
f DeepC
300 1- SAV - 2 300 SAV - a
51 k " 51 gk "
C 0.0 1.0 2.69 30 4.0 5.0 D 0.0 1.0 2.8 30 4.0 5.0
O 1 E 1 1 1 1 1 0 JfE 1 1 1 1 1
‘\, Bhs 1'
?’-1851 \"I 851
11.. 11-. 1/1112
° 852 /
/ . ac
, BC
100 I 100 c
150 «2 5150 «I
g : a :
n 3
200 ~§Deep c 200 ..;
4 DeepC
250 ~- 250 1-
300 -L FOR - 2 300 -- FOR -5

 

Figure 41. Si0 content by depth for four representative pedons. Error bars signify high
and low values in horizons where ortstein was sampled separately.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

125
A FeJFed B FeJFed
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
0 E ,1 1 1 1 1 o E + If Q, j: 4
\.-__' 851 .\P.‘ 851
50 -. \ 50 - l-°/-—| BSZ
, 852 /
/ 0 BC
100 -- ’ BC 100 1 /
. .. .’ C
E E :
.1: 150 -- 5150 -- .
§ . c n- .'
a ' 3 .‘
200 -- 200 g, ..
250 -- 250 -- ;'
:Deep 0
300 -1 SAV - 2 300 -- SAV - 8
C Feolped D FeJFed
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
0 1 1 1 1 1 0 1 1 1 1 1
E 7.8m Bhs E -\v.-t 851
so -- so -- p...
°\ 852 352
, BC i 3°
100 -- \ 100 -- 3 C
‘5‘ '. c ‘5‘ :
9. 1 3 -'
5150 -- .' 4:150 «- -
g , '6. .'
o , " 3
200 "' 0' 'Deep C 200 "" ‘0.
,' DeepC
250 -- 250 1.
300 -- FOR - 2 300 1- FOR - 5

 

 

 

Figure 42. FeO/Fed values by depth for four representative pedons. Error bars signify

high and low values in horizons where ortstein was sampled separately.

126

subsamples (Table 19), suggesting that the pedogenically “active” Fe forms are
concentrated in the ortstein portion of the horizon.

The ratio of inorganic amorphous to organic amorphous Fe ([Feo-Fep]/Fep) is an
indicator of the relative amounts of the non-crystalline Fe that is in organically-bound
forms. It has a maximum value in the upper B horizon, but below the Bhs horizon in
those pedons with a Bhs horizon (Figure 43). A similar profile distribution was reported
in Spodosols of northern Quebec (Wang et al. 1986). The ratio is greater in the matrix
than in the ortstein subsamples of the B51 horizons (Table 18), and is also significantly
higher in stump prairie pedons over forested pedons (Table 17). Values in this study
ranged from below 0 to 10.8 (mean 0.65), a range similar to those reported by Barrett and
Schaetzl (1992) in Spodosols of northern Lower Michigan, but lower than that seen in
northern Quebec (Wang et al. 1986). In the sampled pedons, most amorphous Fe is in
organically-bound form, but inorganic amorphous Fe exceeds organically-bound Fe just
below the Bhs horizon and in non-cemented portions of most upper Bs horizons. This
may imply that Fe has been translocated out of the surface horizons in organically-bound
form, but upon reaching the upper B horizon, it may be re-mobilized and translocated
further in inorganic form.

The ratio of inorganic amorphous to organic amorphous Al ([Alo-Alp]/Alp) shows
a distribution similar to the analogous Fe ratio, with a maximum below the uppermost B
subhorizon in many pedons (Figure 44). As for Fe, a higher proportion of inorganic
amorphous Al than organic amorphous Al is found in the matrix over the ortstein
subsamples of the BSI horizon (Table 19). This suggests that most of the Al and Fe in
the cemented portions of the pedons are organically bound, and that the inorganic forms
of Al exceed organically-bound forms below the organic-rich upper B horizon.

The distribution of inorganic amorphous Al (Ale-Alp), representing poorly
crystalline aluminosilicates like imogolite and/or allophane (Parfitt and Kimble 1989;
Jersak et al. 1995), also shows a maximum in the upper B horizon, below the Bhs in those

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A (Fe,-Fa,)/Fe, B (Fe.-Fe,)IFe,
0.0 1.0 2.0 3.0 4.0 0.0 1.0 2.0 3.0 4.0
1 1 1 1 1 O 1 1 1 1
o ,5
o—1 Bs1 \i-l BS1
50 ._ 50 r- 57". 852
e 852 . BC
100 BC 100 «- \
‘5‘ ‘2‘ 2°
2. .2. 1
5150 -- 5150 q- '.
1:- .c . =
o 3 '.
200 «- 200 -- '1
250 1- 250 -~ I.
‘2 DeepC
300 -- SAV - 2 300 i SAV - s
C (Foo-FepllFe. D (Fe.-Fa,)/Fa
. 0.0 1.0 2.0 3.0 4.0 0.0 . 2.0 3.0 4.0
O 1 1 1 1 0 1 1 1 1 1
”5‘. Bhs 2 E B 1
‘ «4 Bs1 " s
50 1- 5° 1' 1.! 852
852 9%
100 -‘ BC 100 -- ,2 c
.5. .= ° .5.
5150 --; 5150 -- ,-
a- o n '
8 p: 3 .'.
200 J Deep c 200 «1
’Deep C
250 ._ 250 -.
zoo ._ FOR - 2 300 -- FOR .5

 

 

Figure 43. (F eo - Fep)/Fep values by depth for four representative pedons. Error bars
signify high and low values in horizons where ortstein was sampled separately.

 

 

 

 

 

 

 

 

 

 

 

A (AI.-AI,)IAI,, B (AI.-AI,)IAI,
0.0 1.0 2.0 3.0 4.0 o. 1.0 2.0 3.0 4.0
O . 1 1 1 0 ~ E 1 1 1 1
5.4 851 \r-o< 351
50 .. / 50 "' l—-—’—1 852
9 852 ,/
/BC BO
100 .. " 100 -- /
A A 9 C
E E :
3 a .
4: 150 -~ 5150 _. ;
5. C a :
0 ‘.’ 0
a o :
20° 1- 200 -~ ;'
250 -- 250 1- f
. Deep C
300 -- SAV -2 300 -- SAV - 8
(Al -Al )IAI (Al -Al )IAI
C o 1.0 ° 2.5 '’:1.0 4.0 D 0.0 1.0 ° 2.5 "3.0 4.0
O E 1 1 1 1 O ' 1 1 A 1
”\2 Bhs E
A... 351 i1 851
5° T , 852 5° i352
. ac
. BC \
100 «I 100 C
i=3 2° E E
‘5150 «j Z150 -1
3 : § ‘
D : D ‘
200 J Deep C 200 i-
? Deep C
250 -~ 250 --
300 -. FOR -2 300 -- FOR -5

 

 

 

 

 

Figure 44. (A10 - Alp)/Alp values by depth for four representative pedons. Error bars
signify high and low values in horizons where ortstein was sampled separately.

 

129

pedons with Bhs horizons (Figure 45). Very low (< 1.0 g kg']) Ale-Alp contents were
found in some forested pedons (Figure 45D). AlO-Alp content is higher in matrix than in
ortstein subsamples of B31 horizons (Table 19). The depth distributions of inorganic
amorphous Al and the ratios of inorganic amorphous to organic amorphous Al could
indicate translocation below the uppermost B horizon of Al in inorganic amorphous form,
independent of organically-bound Al. Similar depth distributions and values were found
by Wang and Kodama (1986) in pedons in which the presence of imogolite was detected
in the lower B horizons. In this study, I did not examine samples for imogolite directly,
so its presence in the lower B horizons cannot be confirmed.

A further indication of the presence of short-range ordered aluminosilicates in
soils is the molar ratio of inorganic amorphous Al to oxalate-extractable Si ([Alo-Alp]/Sio;
Gustafsson et al. 1995; Takahashi et al. 1995). In the sampled pedons, (Al(,-Alp)/Si0 is
usually at its maximum value in the B horizon (Figure 46). Values of the (Alo-A1,,)/Sio
ratio in the sampled pedons are low (most < 2.0; Table) compared to those reported for
Spodosols in the literature (> 2.0; Evans and Wilson 1985; Parfitt and Kimble 1989;
Dahlgren and Ugolini 1991; Shoji and Yarnada 1991; Gustafsson et al. 1995; Takahashi
et al. 1995). Even though the ratio is low, Ala-Alp is a good predictor Of Sio in these soils
(Sio = 0.20 + O.97[Alo-Alp]; 12 = 0.88), indicating a 1:1 ratio ofA1:Si. Farmer et al.
(1983) reported that the Al:Si ratio in allophanic material is close to 2:1. Inoue and
Huang (1986), however, found that precipitates formed in the presence of organic ligands
with a strong affinity for A1 at a Al:Si ratio of 1.0 are very similar to protO-imogolite, and
are potentially mobile within the soil profile. Fulvic acids have been shown to affect the
genesis of allophane and imogolite such that the Al:Si ratio will increase with increasing
CC (Huang 1991). The (Al,,-Alp)/Alp depth distribution (Figure 44) shows the highest
Al:Si ratio in those horizons with the highest ODOE values (Figure 38). It is possible,
therefore, that the low Al:Si ratios in these soils are a reflection of the amounts and types

of organic matter present in the soils. Allophanes with Al:Si ratios < 2 are probably a

130

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Al -Al k “ Al -Al g kg"
A 0.0 0 2.0 "93.09 4.0 5.0 B 0.0 1.0 2.0 " 3.0 4.0 5.0
01 E 1 1 1 1 1 0 TE 1 1 1 1 1
\ot 851 ’* 351
50 fl / 5° ‘” 7‘ 852
l’ 852 . BC
100 ° BC 100 --/
—~ - z c
.5. 5.
5150 5150 "I
E. a. ;
3 "C 8 :
200 -- 200 «E
250 -- 250 «5
§ DeepC
300 i SAV - 2 3001 SAV - 3
Al -Al k " Al -Al g kg"
C 0.0 1.0 2.0 ”93.1? 4.0 5.0 D 0.0 1.0 °.0 " 3.0 4.0 5.0
0 1 1 1 1 1 1 0 E 1 1 1 1 1
i\E, Bhs ‘15
fin; 351 - lo-t 851
50 " 50 "'3
o 332 l‘ 852
./ ii BC
10011 BC 100130
5. ’° .8. 5
5150 5150 «r
O. a.
8 8 ’
- 200 .. DeepC 200 1-
1. DeepC
250 -- 250 ..
300 1 FOR - 2 300 1 FOR -5

 

Figure 45. Inorganic amorphous A1 (A10 - Alp) content by depth for four representative

pedons. Error bars signify high and low values in horizons where ortstein was
sampled separately.

 

3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A (AI.-AI,)151. B (Al.—Al,)l$l.
0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0
0 1 1 1 1 1 1 0 1 1 1 1 1 1
~ E E
\P-1 851 ‘NI'Bsi
50 -- 50 -- . 352
/' 852 ‘, 8C
100 1 ° BC 100 .. /
.. A 9 C
E E t
3 3 -
a 150 -- 5150 .. 1
3 , c a :
3 3 :
200 1- 200 -- ';
250 -~ 250 --
3 Deep C
300 -. SAV - 2 300 1 SAV - 3
C (AI.-AI,)ISI. D (Al.-AI,)ISI.
0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0
O 1 1 1 1 1 1 0 1 1 1 1 1 1
9 E E B 1
, ’ Bhs :1... s
'— ' 851 |
50 -- 50 -- . 352
/ s2 /
. BC ’\ BC
100 -- \ 100 -- ,9 c
S. .2 C E 5
5150-» - 5150-7
8- : ‘1 r
o ,- 3 :
200 .: Deep c 200 '-- Deep 5
250 -- 250 -.
300 -1 FOR - 2 300 ‘1 FOR - 5

 

Figure 46. (A10 - Alp)/Sio values by depth for four representative pedons. Error bars
signify high and low values in horizons where ortstein was sampled separately.

 

132

mixture of imogolite-like allophane and silica-rich allophane, but usually form in soils
with high Si in solution and at higher pH (>6.0) than is present in the study area soils
(Parfitt and Kimble 1989).

The molar ratio of Fep to A1p illustrates the differences between organically-bound
Fe and Al distributions within the profiles. The slight Pep/Alp minimum values in the
lowest B5 or BC horizon of many pedons (Figure 47) may be an indication of deeper
translocation of organically-bound Al (Alp) than organically-bound Fe (Fep). The B
horizon maximum suggests that Al is more quickly mobilized than Fe in the surface
horizons. Gustafsson et al. (1995) also found that molar FezAl ratios decreased with
increasing depth in Spodosols of northern Scandanavia, with values similar to thoseof
the present study. F e(III):Al molar ratios > 0.5 have been reported to inhibit the
formation of imogolite (Huang 1991). Pep/Alp values are greater in ortstein subsamples
than in matrix subsamples of B51 horizons (Table 19), possibly indicating that Fe is
relatively important in cemented horizons, even though there is more Al than Fe in
absolute terms. In other ortstein soils, Al-organic complexes have been reported to be the
dominant cementing agent (McKeague and Wang 1980; Lee et al. 1988a, b).

The overall picture of the soils in the present study offered by the OC, Fe, and Al
data is that there is a concentration of 0C, especially fulvic acids (as indicated by
ODOE), at the top of the B horizon and especially in cemented portions of Bs
subhorizons. Most amorphous Fe and A1 is present in organically-bound forms, but
inorganic forms become more important below the Bhs horizon and in non-ortstein
portions of the B5 subhorizons. Organically-bound Fe especially is concentrated in the
Bhs and ortstein parts of Bs subhorizons, but a larger proportion of organically-bound Al
than Fe has been translocated deeper in the solum into lower Bs subhorizons.

The data are consistent with a view of podzolization in which Fe and Al, released
through weathering in surface horizons, are translocated into the upper B horizon as

organo-metallic complexes (De Coninck 1980). In the upper B horizon, and in ortstein

133

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A Pep/Al, B Pep/A1,
0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8
0 1 9 E 1 1 1 O 1 1 1 1
’ E
"/8131 ‘{
50 ~- 50 -- 1. 882
e\852 ] 5c
' BC
100 -~ 100 -- \
-"-' 15° " .r: 150 ~~ .'
‘6. a :
s * C s -
200 ~- 200 -- f
250 -- 250 -. f
‘ DeepC
300 _. SAV - 2 300 i SAV - a
0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8
0 1 E 1 1 1 0 1 1 1 1.
2' e Bhs , ' E
'7' 851 ' 851
50 w o 852 50 “ 7—' 832
/ . BC
. BC I C
100 «- \ 100 -- z
E 9 C E :
9— .' 9.— i
5150 ~- . 5150 -- .
a .' a. I
3 g s g
200 '1“ 0. Deep 0 200 “F I
§ DeepC
250 -- 250 ..
300 -- FOR - 2 300 ~- FOR - 5

 

Figure 47. F ep/Alp values by depth for four representative pedons. Error bars signify
high and low values in horizons where ortstein was sampled separately.

 

134

columns in B5 subhorizons (which presumably indicate pathways of preferred water
flow), the organically-bound forms of Fe and A1 are deposited, accounting for the
predominance of organic Fe and A1 forms in these horizons. A possible re-mobilization
of Fe and A1, with a preference for Al, causes the further downward migration, in
inorganic form (possibly “proto-imogolite sols” sensu Anderson 1982), out of the Bhs
and ortstein subhorizons and into the lower Bs horizons, and deposition there as short-
range-order aluminosilicates. This scenario is consistent with the theory presented by
Dahlgren and Ugolini (1989) for tephritic Spodosols, in which they proposed two
“compartments” of soil processes, the upper compartment (A, E, and Bhs horizons)
dominated by organic acid weathering and translocation of organo-metallic complexes,
and the lower compartment (Bs, BC, and C horizons) dominated by H2C03 weathering in
equilibrium with the hydroxy aluminous interlayer of 2:1 layer silicates and imogolite.
Some of the inorganic Al and Fe in lower Bs horizons may also result from incongruent
dissolution of minerals in situ due to H2CO3 weathering (U golini and Sletten 1991).
From soil morphology and chemical data alone, it is unclear how soil
development processes have been affected by the change in vegetation that occurred with
the establishment of the stump prairie. The primary difference between the forested
pedons and stump prairie pedons is the amount of ortstein present in the soils.
Presumably, greater amounts of organic acids in the litter of forested pedons contributed
to the formation of ortstein and Bhs horizons in these pedons, but it is impossible to
distinguish whether current vegetation patterns or pre-disturbance forest types had a
greater part in its formation. It does not seem likely that approximately a century of
lowered organic acid production in the stump prairie soils could have led to ortstein
disintegration to the degree to account for the current differences in morphology.
Because allophane and imogolite are particularly unstable in the presence Of
organic acids (Inoue and Huang 1986), upper Bs horizons could be easily transformed

into Bhs horizons if the supply of organic acids were to increase (Gustafsson et al. 1995).

135

Thus, a mechanism exists for the development of the spodic horizon and an increase in
ortstein and Bhs content in these soils if the production of organic acids increased
following logging disturbance. It seems unlikely, however, that organic acid production
in forested pedons would have increased; rather, a decrease would be expected in stump
prairie pedons. Whether the reverse mechanism is possible, however, remains unclear.
That is, it is unknown whether a decrease in the supply of organic acids would lead to the
decomposition of already deposited organo-metallic complexes and favor the formation
of amorphous aluminosilicate-rich Bs horizons, and the disintegration of ortstein which

appears to have occurred in the stump prairie pedons.

136

The properties of the solid phase of the soil can provide insight into the processes
that have contributed to soil formation. Because soil properties are determined by the
cumulative effects of all processes active during soil formation, some properties may be
relict features of processes no longer active. Relict soil features may reveal that the soil
is polygenetic due to changes in climate or vegetation (Bryan and Albritton 1943;
Johnson and Watson-Stegner 1987; Ugolini et al. 1987; Ranger and Nys 1994). Studying
the sgiLsglnfiQn can help to identify those processes still active and identify soil
properties that may be relict (Ugolini et al. 1987; Righi et al. 1990). Often lysimeters are
used to study the soil solution by direct analysis (e.g., Grier 1975; Singer et al. 1978;
Ugolini et al. 1988; Schaetzl 1990). In the present study, bags of chelating resins and
cation exchange resins were buried in the soil to deduce the movement of various cations
in the soil solution (Righi et al. 1990; Ranger et al. 1991; Ranger and Nys 1994). The
advantage of using resin bags over lysimeters to study soil solution processes is that the
resin bags require much less on-going maintenance than the lysimeters, allowing study of
pedons in relatively inaccessible sites. It also more readily allows replication of the tests
over space, requiring the installation only of the bags, and not a complex lysimeter
system. The purpose of this part of the study was to determine whether active soil
forming processes in the forest soils were different from those in the stump prairie soils,
especially with respect to the forms and amounts of Fe and A] being translocated (see pp.
132 - 135).

For both the cation exchange resin and the chelating resin, absolute amounts of
sorbed cations were very small (Tables 25 and 26, Appendix A) compared to results
reported by Ranger et al. (1991) in France. The low cation capture totals may be due to
the short period of burial (< 10 months) or to the small amounts of percolating soil

solutions due to low precipitation amounts (43.3 cm actually recorded at Munising by the

137

National Weather Service cooperative observer vs. 50.9 cm for the 30 year mean at the
same site) during the burial period (Ranger et a1. 1991). Nevertheless, despite the small
absolute values, relative amounts of sorbed cations, by horizon, appear to reflect the soil
processes.

Amounts of Mg and Ca sorbed on the cation exchange resin are generally greatest
at the top of the E horizon, with the exception of Ca in the stump prairie pedons, where
top of the B horizon values are higher (Figure 48). Forest pedons usually showed larger
amounts of both Mg and Ca sorbed at the top of the E horizon than stump prairie pedons,
probably due to greater cation recycling by the forest vegetation and release from the
decomposition of the litter layer (Ranger and Nys 1994). For most horizons, amounts of
sorbed Ca were higher than sorbed Mg values. Very little Ca or Mg (Ca mean 76 mg kg'1
dry resin; Mg mean 26 mg kg’1 dry resin) is present in soil solutions reaching the top of
the BC horizon, implying that the lower part of the B horizon is dominated by inorganic
processes and carbonic acid weathering (U golini and Dahlgren 1987) .

In both forested and stump prairie pedons, and both resin types, sorbed Ca was
four to six times higher than sorbed Mg (Table 20). Relative amounts of sorbed Ca and
Mg are in agreement with measurements of cation cycling in a forested ecosystem in New
Hampshire (Likens et al. 1977, p. 101). In Alfisols of northern Minnesota, CazMg ratios
in the forest floor and in mineral soil horizons ranged from 6 to 10 (Alban 1982). The
ratio of Ca to Mg in the biogeochemical cycles of both spruce and broadleaved stands in
France also were determined to be approximately 6 (Ranger and Nys 1994).

Amounts of sorbed A1 are generally higher than sorbed Fe for both the cation
exchange resin (Figure 48) and the chelating resin (Figure 49). Most of the Fe and Al in
the soil solution appears to be in organically complexed forms because the amounts of Fe
and A1 sorbed on the chelating resin are much higher than amounts sorbed on the cation
exchange resin. For most pedons, sorbed Fe and Al on both the cation exchange resin

and the chelating resin are at a maximum at the top of the B horizon. This depth pattern

138

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A Cation Exchange Resin Ca 3 Cation Exchange Resin Mg
mg kg" dry resin mg kg" dry resin
0 500 1000 1500 2000 0 100 200 300 400
FOR-5 ' FOR-5 E
1.
FOR-7 ' FOR-7 ‘
FOR-9 FOR-9
SAV-o SAV-o
SAV.7 DTop of E horizon ”v.7 DTop Of E horizon
ITopothorizon ITopoiBhorlzon
SAV-a sav-e
DTop of ac DTop at BC
horizon horizon
c Cation Exchange Resin Al D Cation Exchange Resin Fe
mg kg" dry resin mg kg“ dry resin

0 100 200 300 400

1
I

 

 

  

V 1

FOR-5 FOR-5

F OR-7 FOR-7

FOR-9

 

FOR-9

SAV-B SAV-B

 

 

 

SAV-7 saw 0 TOP of B horizon
I Top of B horizon
sAV-a 5Av-a
O Top of ac
horizon

 

 

 

 

 

 

 

 

 

 

Figure 48. Amounts of cations sorbed on cation exchange resins for the six sampled
pedons. Error bars indicate minimum and maximum values from replicate bags.
Note that part A has a different axis scale than other parts. A) Ca; B) Mg; C) A1;
D) Fe.

139

Table 20. Mean amounts of cations sorbed on cation exchange and chelating resins in
forest and stump prairie pedons.

 

Forest pedons“ Stump prairie pedons“

 

Wm

 

 

Cation exchange resins (all horizons)

Mg 62.7 30.5
Ca 226.2 87.4
Fe 30.1 18.7
A1 93.9 37.3

Chelating resins (all horizons)
Fe 1 17.3 64.0
Al 473.4 282.7

Cation exchange resin (top of B horizon only)

Mg 65.5 36.9
Ca 155.3 142.3
Fe 60.8 27.6
Al 201.4 59.6

Chelating resin (top of B horizon only)
Fe 255.6 130.9
Al 959.0 550.7

 

*F or all rows, column means are not significantly different (p = 0.05, Kolmogorov-
Smimov).

140

 

A

FOR-5

FOR-7

FOR-9

SAV-B

SAV-7

SAV-B

 

   

Chelating Resin Al
mg kg" dry resin
0 500 1000 1500

J L
Y

 

DTopoiEhonzon
ITopoiBhorizon

DTopofBC
horizon

 

 

 

 

 

FOR-5

FOR-7

F OR-9

SAV-B

SAV-‘I

SAV-8

Chelating Resin Fe
mg kg" dry resin
500 1000 1500 2000

l L
Y I T j

 

DTopoiEhorizon
ITopoiBhorizon

DTOpOfBC

 

 

 

 

 

Figure 49. Amounts of cations sorbed on chelating resins for the six sampled pedons.

Error bars indicate minimum and maximum values from replicate bags. A) A1; B)

Fe.

141

suggests that amounts Of Fe and Al leaving the litter layer in the soil solution are
relatively small, and, therefore, that the source of soil solution Fe and Al is weathering in
the E horizon. The soil solution leaving the B horizon also has very low Fe and Al
content, suggesting that most of the Fe and Al is immobilized within the B horizon

(U golini et al. 1977; David and Driscoll 1984; Schaetzl 1990; Ugolini and Sletten 1991).

Although not statistically significant, mean amounts of sorbed Fe and Al appear to
be slightly greater in the forest pedons than in the stump prairie pedons (Table 20).
Variability both within and between pedons is very high, however. Pedon F OR-7,
located in a relatively undisturbed forested area with little evidence of recent logging and
larger trees than most other forest in the area, consistently shows the largest amounts of
all sorbed cations; pedon SAV-8 is consistently low (Figures 48 and 49). If greater
amounts of cations have been sorbed on resins in forested pedons, this has happened
despite the fact that less water typically moves through soils in forested than in
unforested locations due to interception by tree crowns (Schaetzl and Isard 1990). From
the data available it appears that podzolization processes may be more active in the forest
than on the stump prairie. Nevertheless, podzolization seems to be an on-going process
in the stump prairie soils as well.

The patterns of Fe and Al sorbed on the chelating resin are typical of patterns
found in studies of soil solutions using lysimeters in podzol soils, in which Fe and Al
concentrations in soil solutions are low just beneath the litter layer, highest as the solution
enters the B5 horizon, and low again as it leaves the horizon. Such patterns have been
found in Michigan (Schaetzl 1990), the Arctic (U golini et al. 1987), Washington (U golini
et al. 1977; Ugolini and Sletten 1991), New York (David and Driscoll 1984), and
northern Japan (U golini et al. 1988). It is likely, therefore, that the cations sorbed on the
resins are qualitatively representative of the cations moving in the soil solutions in the

study area soils.

142

The use of resin bags to study soil solutions has been employed to identify a soil
in which active soil processes did not match morphology, suggesting a relict soil (Ranger
et al. 1991), and also to distinguish soil processes resulting from differences in vegetation
types (Carlyle and Malcolm 1986; Ranger and Nys 1994). In the present study, soil
processes are in accord with soil morphology, and podzolization processes appear to be

more active in the forest than in the stump prairie soils.

Summary

In summary, evidence from soil properties and an in situ examination of soil
processes using resin bags suggests that the soils in the forest and the stump prairie are
similar, but distinctions between the two groups do exist. Parent material texture is the
same in both groups, but amounts of Fe, Al, and 0C are higher in the forested than stump
prairie soils. Most differences between the two groups of soils can be attributed to
differences in ortstein content; forest soils have more ortstein than stump prairie soils,
and ortstein samples typically have higher Fe, Al, and OC contents than matrix samples.
Thus, the spatial relationship between forest regeneration patterns and soil patterns is best
expressed in ortstein content of the soils.

Depth distributions of Fe, Al, and DC in both forest and stump prairie soils are
consistent with current theories of podzolization, showing translocation of Fe and A1 in
organically complexed form in upper horizons, and possible further translocation of
inorganic forms of Al primarily into lower horizons. Soil solutions, studied using resin
bags, indicate that podzolization processes are active in both groups of soils but appear to
be more so in the forested pedons. NO evidence was found for the notion that differences
in degree of soil development could be attributed to processes destroying the spodic
horizon due to the change in vegetation type from forest to grassland in the stump prairie

region.

143

Factor 3: Logging era fires and logging practices

Logging

No one questions that disturbance associated with nineteenth century logging has
led to the establishment of the stump prairie. The general history of logging in the Upper
Peninsula and on the Kingston Plains has been discussed above (pp. 13, 21). Whether
logging practices in the areas that today are stump prairie were different from those where
the forest has recovered, however, remains unclear. This section will examine historical
records relating to the logging of the study area for evidence of whether logging practices
in the two current landcover classes differed in either (1) character or (2) timing.

Specific records of dates and methods of logging at the end of the last century are
difficult to obtain. As a surrogate, more easily-obtainable records pertaining to land
ownership and property tax payment have been utilized in this study to establish who
owned the land during the years the area was being logged, and the assessed value of that
land. Two assumptions have been made relative to these records: (1) most logging that
took place on the Kingston Plains was authorized by the owners of the land; and (2) the
assessed value of a land parcel with standing, marketable timber was much higher than its
value after it had been stripped of its lumber. Since illegal logging by people other than
the land owner was not uncommon in nineteenth century Michigan (Skeels 1898), it is
possible that the first assumption does not strictly hold. Nevertheless, lands owned by
large lumber companies were most likely logged by that company.

mm The land in township in which the study area is
located was purchased from the government over a period of about 20 years, with most
activity taking place before 1880. This period is considerably before logging in the

central Upper Peninsula was common, and well before any logging began in the vicinity

144

of the Kingston Plains. The original purchasers can be divided into five main groups: (1)
E. B. Ward; (2) the Sault Canal Company; (3) the Lac La Bell Harbor Improvement
Company; (4) S. Q. Perry and associates; and (5) other individuals with smaller holdings
(Figure 50). The land ownership patterns established in this period persisted through the
logging era. For the most part, however, the actual ownership changed before logging
took place. The spatial pattern of the dates of acquisition from the government reflect
land ownership patterns because the major purchasers acqired their land at different times
(Figure 51).

The first acquisition of land in the study area from the government was in 1855 by
Eber B. Ward, a Detroit, Michigan, industrialist, later associated with the steel industry
(Burton and Burton 1930). Mr. Ward’s purchases were concentrated in, but not limited
to, the areas that today are stump prairie. Also in 1855, the Sault Canal Company
obtained land in the southeastern portion of the township, which is generally wet and
forms the headwaters of a river. This land may have been a grant to the Sault Canal
Company in exchange for canal improvements (Hibbard 1939, p. 238). Similarly, the
odd—numbered sections of the township not already claimed were granted to the Lac La
Belle Harbor Improvement Company in 1866. After 1870, acquisition by individuals,
probably as lumber speculation, became common. Silas Q. Perry and a number of his
associates, who later organized the Manistique-based North Shore Lumber Company
(Hotchkiss 1898), acquired many parcels beginning in 1872. Mr. Perry’s parcels were
located both in stump prairie and in forested areas.

If forest composition of the stump prairie areas was of markedly greater quality
for logging purposes because of its higher proportion of white pine, it might be expected
that the stump prairie land should have been acquired before adjacent regions. Certainly
E. B. Ward’s early acquisition seems to follow that pattern (Figure 50). The fact that the
Lac La Belle Co. was granted all odd-numbered sections in the 1860’s, however, makes it

difficult to discern whether this pattern continues in later purchases, because little stump

 

E. 8. Ward

% S. Q. Perry

E Lac La Belle CO.
[HIM] Sault Canal CO.
[:1

Other

Figure 50. The original purchaser of land from the government, T 48 N, R 15 W, Alger
County, Michigan. The stump prairie/forest boundary is shown for reference.

146

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1850 - 1859
1860 - 1869
[fill] 1870 - 1879

[:1 1880 - 1889

Figure 51. The date of the original purchase of land from the government, T 48 N,
R 15 W, Alger County, Michigan. The stump prairie/forest boundary is shown
for reference.

147

prairie land remained for sale after 1870. Those parcels acquired last, after 1880, were
primarily located in areas that are today forested.
WWW Tax assessment rolls show that the
owners of the land in the township remained largely unchanged at least through the year
1887. A few parcels defaulted to the state for non-payment of taxes in the late 1880s,
particularly some of S. Q. Perry’s land (Abstracts of sales of state tax lands, Alger
County, 1888, 1890, 1893). Between 1887 and 1890, however, the ownership of the land
in the area underwent a major change (Figure 52). S. Q. Perry’s land, including that
previously defaulted on, was acquired by Hall and Buell, another Manistique-based
lumber company (Hotchkiss 1898). Most of the Sault Canal Company’s land and the
land belonging to E. B. Ward was acquired by the Manistique Lumber Company, of
which R. A. Alger, a civil war general and governor of Michigan from 1884 to 1888, was
one of the major stockholders (Bell 1975, p. 134). The Lac La Belle Harbor
Improvement Company apparently retained ownership of most of their land, but in the
1890 tax rolls its ownership is listed as “unknown” and the taxes were unpaid. The
Alger-Smith Company (also belonging to R. A. Alger) did acquire some of the Lac La
Belle land before 1890 (Figure 52). By 1890, therefore, most of the land that is today
stump prairie had been acquired by one of two lumber interests: (1) R. A. Alger (the
Manistique Lumber Company and the Alger-Smith Company) and (2) Hall and Buell.
The holdings of these two companies were not entirely restricted to stump prairie land.
The assessed land value in 1890 generally ranged from $4.00 to around $7.00 per
acre (Figure 53). Patterns of assessed value appear to be more closely related to the land
ownership than forest patterns; e. g., most of the Manistique Lumber Company’s land was
valued at $7.00 per acre, but Lac La Belle’s land was assessed at $4.00 per acre. From
the assessed land values shown in Figure 53, it can be surmised that little logging had
taken place in the township by 1890. Of interest here is the small parcel of land assessed

at $1.00 per acre and owned by one of the minor landholders (Figure 53). It is possible

148

     

\ h

Alger-Smith CO. or Mani5tique Lumber CO.

 

% Hall and Buell

[:1 Other

Figure 52. Land ownership in 1890, T 48 N, R 15 W, Alger County, Michigan. The
stump prairie/forest boundary is shown for reference.

 

1:] $1.00 or under
E $1.00 - $4.00

7% $4.00 - $6.00

 

Over $6.00

Figure 53. Per-acre assessed land value in 1890, T 48 N, R 15 W, Alger County,
Michigan. The stump prairie/forest boundary is shown for reference.

150

that this represents an area that was logged early. The parcel is located adjacent to, but
not entirely within, a stump prairie area.

Between 1890 and 1895 land ownership in the study area remained essentially
stable (Figure 54), but major changes in assessed value had taken place (Figure 55).
Most parcels that had been worth more than $4.00 per acre in 1890 were assessed at
$1 .00 per acre in 1895. Assessed valuation in 1895 is much less closely tied to land
ownership than it had been in 1890, probably reflecting a major devaluation due to
logging activity that occurred between 1890 and 1895. Of interest is the appearance on
the tax rolls in 1895 of State Tax Land (Figure 54), all of it formerly belonging to Hall
and Buell (Figure 52). The State Tax Land was uniformly assigned a per-acre value of
$0.50. It was scattered among Hall and Buell’s holdings, both within and outside regions
that are stump prairie today. By 1899 the State Tax Land had been redeemed by either
Hall and Buell or the Manistique Lumber Company. Most was still valued at $0.50 per
acre.

If a per-acre assessment of $1.00 per acre or less is taken as evidence that a parcel
had been logged, it is apparent that widespread logging took place between 1890 and
1895 in both forested and current stump prairie regions. Some stump prairie regions
remained unlogged in 1895, but were probably logged shortly thereafter. By 1899, most
of the township was assessed at $1 .00 per acre or less (Tax assessment rolls, Alger
County, 1899), with the major exception being the the swampy, Fox River headwaters in
the southeastern comer of the township. Land ownership was largely unchanged. In
1899, the Manistique Lumber Company is recorded as having paid the taxes on its
holdings, but the Hall and Buell parcels were again unpaid.

In conclusion, the tax assessment and ownership data suggest that most logging
activity on the Kingston Plains took place between 1890 and 1895, and was virtually
completed by 1899. These tax records show that land ownership patterns in the study

area were constant from the time of its original acquisition through the end of the century.

151

     

\ h

 

Alger-Smith CO. or Mani5tique Lumber CO.
Hall and Buell
Lac La Belle CO.

State Tax Land

E
.\\V
I:

Figure 54. Land ownership in 1895, T 48 N, R 15 W, Alger County, Michigan. The
stump prairie/forest boundary is shown for reference.

Other

 

El $1.00 or under
E 51.00 - $4.00
$4.00 - $6.00

 

Over $6.00

Figure 55. Per-acre assessed land value in 1895, T 48 N, R 15 W, Alger County,
Michgan. The stump prairie/forest boundary is shown for reference.

153

Logging activity in the township was primarily carried out by two separate logging
interests, the Manistique Lumber Company and Hall and Buell. The holdings of these
two companies are intermixed throughout both the stump prairie and the forested areas,
but appear to be slightly concentrated in the stump prairie area. It is unlikely that one
company’s logging practices were more detrimental than those of the other to the
recovery of the forest. The minor landowners, however, whose holdings were located
where it is now forested, might have had distinct logging practices that could have
contributed to the recovery of those parcels. The apparent date of logging does not seem,
from these data, to be associated spatially with the forest recovery pattern, as logging of

both forested and stump prairie tracts occurred within the same time period.

Fires: Evidence for burning

Careful Observation in the stump prairie portion of the study area, as well as in the
forests, provides an unmistakeable impression that much of the area has burned. Many of
the old stumps contain charcoal, both in the forest and in the stump prairie. Therefore,
this section will examine whether there were differences in either frequency or intensity
of burning between the two areas.

Historical discussions of logging in the area commonly mention fire. A short
description of the Kingston Plains, apparently written for game managers around 1940,
states that the former superintendant of the Cusino C. C. C. camp claimed that a crown
fire occurred while the timber was being taken out, and necessitated a speed-up in
operations to salvage the standing burned timber. The description claimed that the last
fire in the area must have occurred before 1910 because a few widely spaced pine trees
were found to be about 30 years old in 1940 with live limbs down to the ground,
indicating that they had not been burned. The only other recorded fire that took place in

August, 1936; it burned part of the eastern portion of the Kingston Plains.

154

Specific information about the fires that took place is difficult to obtain. The
Kingston Plains area is distant from any permanent settlement even today, and fires were
so common during the logging era that many must have burned, unknown to local
inhabitants. Therefore, it is difficult to establish whether there was a difference in
frequency or intensity of burns between the stump prairie and the surrounding forested
areas. Some surrogate evidence indicating the general nature of burning patterns,
however, is available.

Charm. Charcoal found in the soil can indicate whether the site has
experienced burning, but it does not indicate the intensity of the burn. Once formed, it
may persist in the soil for centuries, so its presence in any of the sampled soils does not
necessarily indicate that burning took place following nineteenth century logging. Of the
pedons sampled, all except two contained at least some charcoal in the surface horizons.
The two that did not are FOR-2 and FOR-7. FOR-7 is located among a stand of
particularly large trees, including white pine and hemlock, which also show no evidence
of having been burned (see below). It is possibly FOR-2’s location near a lake (Figure 3)
shielded it from burning. Extensive evidence of charcoal in soils of forested sites
indicates that most did burn, at least lightly, at some time in the past.

Mags. Perhaps the best evidence for differences in burning patterns between
forest and stump prairie can be obtained from the trees and stumps themselves. The ages
of living trees can be established by means of ring counts. A living tree may have
survived a fire, although some species are more susceptible to fire injury than others, and
a severe fire should at least have lefi a scar on the tree. Thus the presence of a tree with
no scars, especially a fire-susceptible tree, older than about 100 years, would suggest that
that region had escaped severe fire. Similarly, stumps from recent logging activity can be
examined for ring counts and fire scars. In this section evidence obtained from tree

increment boring and stump ring counts will be compared to general field observations,

155

to address the question of whether fire affected the stump prairie more severely than the
forested areas.

Evidence of this type must also be considered with logging patterns in mind. The
lack of surviving trees older than about 100 years in an area may have resulted from
burning, or it may have been the result of a complete removal of all trees by the loggers.
Thus, differing logging patterns would also contribute to differences in tree survival. The
question is, (1) did the loggers leave any living trees, and (2) did the trees that were left
survive the fires which followed the logging?

In the areas where the forest has regenerated, abundant evidence exists, despite
modern logging efforts, that at least some trees remained alive through the time of
logging and the fires that followed, including some white pine trees (Table 21).
Apparently in these areas the loggers left some larger trees, considered inaccessible or
unsuitable, as well as trees too small to be valuable for lumber.

A few very large white pine trees (> 80 cm DBH), located along with large trees
of other species, were encountered in stands FOR-7 and FOR-8, though their age was not
determined. While it is possible that these trees could have sprouted following
lumbering, their growth form suggests that they grew under closed-canopy conditions.
Similarly, very large stumps in stand F OR-4, both white pine and hemlock, remain from a
selective cutting long enough ago for decay to have begun, although bark remains intact.

The best evidence that not all trees were removed during the original logging,
however, comes from stumps left following recent logging efforts. Stand FOR-l was
logged in the spring of 1995, and an area near the border of sections 29 and 30 at the
southwestern comer of the stump prairie was logged in the fall of 1994 or spring of 1995.
Many large white pine and hemlock trees were removed from stand FOR-l , some of
which had nearly 200 annual rings, although some stumps of similar diameter had fewer
than 80 rings. Many hardwood trees in the stand also had well over 100 annual rings.

Prior to this recent logging, large hemlock stumps from a logging episode approximately

156

Table 21. Representative ages from ring counts of selected large trees in sampled stands.

 

 

Stand Species Evidence Diameter AgeI Comments
type (cm)
FOR-1 White pine tree 87 75
White pine, Hemlock stumps 80-100 200 Logged spring 1995; no scars
White pine, Hemlock stumps 80-100 80 Logged spring 1995; no scars
FOR-2 Beech stump 50 120 Logged 10 yrs ago
White pine stump 67 67
White pine stump 50 74
White pine stump 87 87
Hemlock stump <50 1 97
FOR-4 Beech tree 40 Long scar on trunk
White pine, Hemlock stumps 80 ? Largest trees cut approx. 30
years age
FOR-5 White pine tree 75 72
F OR-6 White pine stump 85 Logged approx. 10 years ago
FOR-7 White pine tree 99.5 ? Not open grown
F OR-8 White pine tree 84 ? Soar on side of trunk
SAV-l White pine tree 70 43 On side of railroad grade
White pine tree 30 32 Probably descendent of above
tree
White pine tree 87 44 Double-stemmed
White pine tree 83 61
SAV—2 White pine tree 30 24 Growing in old stump
Black cherry tree 14 25
Red maple tree 21
SAV—3 White pine tree 81 69 Many smaller white pine
growing near it
White pine tree 22 20
Black cherry tree 1 1 18
SAV-4 Almost no trees due to
controlled burns
SAV-6 White pine stump 6O 56 Logged approx. 10 years ago
White pine tree 81 61
Sec.30 Beech, Red maple stumps 60-80 120 + Logged 1994; no scars

 

lRings counted at breast height (1.4 m) or at the top of the stump.

157

30 - 40 years previous had been observed, suggesting that a selective cutting of large trees
had already occurred in that stand. Many red maple and beech stumps in section 29 had
well over 100 rings; the trees had a diameter of 15 - 20 cm at the time of logging 100
years ago. A hemlock stump with 197 rings was also Observed near stand F OR-2.

On the other hand, no conclusive evidence could be found that any tree on the
stump prairie remained after nineteenth century logging operations (Table 21). Very
large, isolated white pine trees exist on some parts of the stump prairie, but these trees are
open-grown with many branches along the lower trunk, and apparently grow rapidly. In
many cases the largest trees on the stump prairie are growing on the disturbed soil (small
borrow pits or cuts) along the sides of former logging railroad grades, and therefore must
have sprouted at or following the time of logging. Some large white pines on the stump
prairie were harvested about 10 years ago, especially in the vicinity of stand SAV-6 and
in a stump prairie located in the southwestern part of section 8. All of these trees had
grown to diameters of 60 - 80 cm or more, and all were less than 90 years old.

The fact that no tree from before the time of logging could be found on the stump
prairie, in contrast to the ample evidence of pre-logging trees remaining in currently
forested areas, suggests that the treatment of the stump prairie during nineteenth century
logging was distinct from that of the surrounding areas. Apparently the white pine stand
located in the stump prairie was pure enough that nearly all trees were cut in the initial
logging. If small white pine trees and trees of other species and were left by the loggers,
they must have been killed either in the fires that followed or due to the altered
microclimate resulting from the sudden removal of the canopy trees.

None of the recently logged stumps showed evidence of scarring due to fire,
whether on the stump prairie or in the forest. Living trees in the forest do not have
obvious fire scars either. Logging-era burned pine snags and stumps are encountered,
however, even in forested areas, suggesting that some fire must have passed through at

least portions of the forests. Some 120+ year-old beech and red maple stumps with no

158

fire scarring found in some recently logged stands indicates that these places either did
not burn or did not burn intensely following logging. If many more trees were left
standing in regions that are today forested than on the stump prairie, it could have
provided shade for regeneration and helped to conserve moisture and prevent the intense

conflagrations and repeated burning that must have occurred in the stump prairie regions.

SUMMARY AND CONCLUSIONS

The forest and stump prairie patterns that exist in the Kingston Plains area are
remarkable, especially given that they have persisted since the logging and disturbances
of the late 19th century. In this dissertation I examined the spatial patterns of a number of
factors, both physical and anthropogenic, that may have influenced the formation or
maintenance of the stump prairie. The study focused on three factors: (1) pre-
disturbance forest patterns; (2) substrate and soil properties; and (3) logging era fires and
logging practices. A secondary purpose of the research was to explore the relationship
between forest vegetation and soil development (podzolization) processes in sand soils of
northern Michigan, and to examine whether, in fact, the long-term elimination of the
forest has altered the pathways of soil development, or even caused soil degradation.
Thus, the study has been organized around the assumption that both physical processes
and human actions have had important influences on the spatial patterns of forest
regeneration in the region, and, in fact, that specific conditions of pre-existing forest type,
soil development properties, and logging practices were pivotal in the establishment of
the stump prairie.

The pre-disturbance forest that existed in current stump prairie areas was
considerably different from the forest that existed where it has regenerated, as shown in
the General Land Office survey notes. In particular, the forests in the stump prairie areas
were predominantly pine forests, with a major component of white pine. Especially
notable is the near lack of any record of sugar maple trees in the areas that today remain
stump prairie, and its prevalence in areas that today are upland forests. Current upland
forest areas were occupied primarily by beech-sugar maple-yellow birch forests, which

also contained some white pine and hemlock.

159

160

Evidence from soil properties suggests that the soils in the forest and the stump
prairie are similar, but subtle distinctions between the two groups do exist. Parent
material texture is indistinguishable between the two regions, but amounts of Fe, Al, and
0C are higher in the forested than stump prairie soils. Most differences between the two
groups of soils can be attributed to differences in ortstein content; forest soils have more
ortstein than stump prairie soils. Thus, the spatial relationship between forest
regeneration patterns and soil patterns is best related to the ortstein content of the soil.

The distributions of Fe, Al, and OC within the profiles of both forest and stump
prairie soils are consistent with current theories of podzolization, suggestive of
translocation of Fe and Al in organically complexed form in upper horizons, and possible
further translocation of inorganic forms of Al primarily into lower horizons. Soil
solutions, studied using bags of cation exchange and chelating resins, indicate that
podzolization processes are active in both groups of soils but appear currently to be
stronger in the forested pedons. Little evidence was found to suggest that differences in
degree of soil development could be attributed to processes destroying the spodic horizon
due to the change in vegetation type from forest to grassland in the stump prairie region.

Tax assessment and ownership data show that land ownership patterns in the
study area were constant from the time of its original acquisition in the 18605 and 18705
through the end of the century. Logging activity was primarily carried out by two
separate logging interests, and the holdings of these companies were centered on the
stump prairie areas, suggesting that the stump prairie areas may have received somewhat
different treatment than the surrounding areas when they were logged. The date of
logging, in the early 18905 for the entire region, was apparently not associated spatially
with the forest recovery pattern. Based on the presence of burned stumps and snags, as
well as charcoal found in the soil, logging-era fires probably affected both forested and

stump prairie areas to some degree. Unscarred trees in some currently forested areas

161

dating from well before the logging era suggest that fires may have been less severe and
logging efforts less complete in forested regions than in adjacent stump prairie areas.

The patterns of soils, pre-disturbance forests, logging practices, and fires suggest
that both physical and anthropogenic factors have a spatial correspondence with the
pattern of forest recovery in the Kingston Plains area today. No differences in substrate
texture between forests and stump prairie were observed, but (the distinctions in the pre-
disturbance forest of the two regions probably led to differences‘irtlogging practices and
fire frequency and intensity that precluded the rapid return of forest species in the stump
prairie. Observations from historical aerial photographs show, however, that the
character of the stump prairie is slowly changing, such that more trees, especially white
pine trees, are present today than in the 19305. It seems likely that, left undisturbed, the
stump prairie areas will slowly recover, and with the passage of time, a pine forest could
become established there.

The Kingston Plains stump prairie presents a unique opportunity for studying the
adverse effects of human activity on the landscape. At first glance, the obvious change in
vegetation type on the stump prairie would seem to point to gross mismanagement of the
land by the lumber companies in the late nineteenth century. What this study has shown,
however, is that specific natural conditions unique to this site also played a part in the
development of the stump prairie. The forest that existed on the Kingston Plains in the
late nineteenth century was rich in white pine, a species that the lumber companies
sought. The concentration of white pine invited the loggers to cut the forest there
especially thoroughly. In the cultural climate of the time, the nearly inevitable
consequence of this type of logging activity on xeric upland sites was frequent and
intense fire. The combination of fires and dry, infertile soils meant that conditions were
unfavorable for forest regeneration, and the result was a stumped landscape of rolling

sand plaines, lichen, and bracken fern. It is likely, however, that if any of the links in this

162

chain of conditions had been lacking, forest regeneration pathways in the area would have
been quite different.

The Kingston Plains is an example of a site where the spatial coincidence of
natural factors and human actions led to an abrupt change in the ecosystem. Although
human actions were the impetus for the changes that occurred, the physical characteristics
of the stump prairie site itself were prerequisite for catastrophic change due to logging.
Thus this site was especially vulnerable (“environmentally susceptible”; Quinn 1991) to
injury from this specific human activity. The aim of this dissertation has been to recount
what happened in the Kingston Plains, and to investigate why it happened there, and not
elsewhere. The Kingston Plains is, therefore, not simply an example of lack of
environmental sensitivity on the part of greedy lumber companies. The Kingston Plains,
like many cases of environmental degradation, represents the results of an interaction
between the natural environment and human actions.

Recognition that a particular location may be environmentally susceptible does
not negate or relieve the human responsibility for environmental degradation. Rather, it
should remind us that in land management decisions we are responsible for understanding
as much of the total landscape as possible before we undertake actions which could upset
the balances that exist. Especially in sites which may be more ecologically fragile, such
as areas prone to drought, it is imperative to attempt to determine the scope of our
potential impact before undertaking wide-scale land management schemes. It is possible
that a stump prairie might not have developed in the Kingston Plains if the logging
disturbance had been accomplished in a “gentler” manner, even given the lack of fire
controls of the time: i.e., using a selective cut rather than a clear out. A selective cut
might have retained enough trees for shade to allow forest regeneration to occur at a
much faster rate. Granted, selective cutting was not the typical method in which logging

was accomplished at the time. Similarly, we should be looking to see if our “usual” land

163

management practices could be harboring an unintended threat to environmentally

susceptible parts of the landscape.

Suggestions for future research

This dissertation has investigated spatial patterns of forest vegetation, logging,
and soils in order to elucidate the factors that might have contributed to the origin and
maintenance of the stump prairie in the Kingston Plains. Although the spatial
coincidence of many of these factors is remarkable, causal mechanisms have not been
established. Further investigation of the causal mechanisms involved with tree
regeneration following fire in the soils of the area should be undertaken.

The logging history of the Kingston Plains, and, in fact, of much of the Upper
Peninsula, has received little attention from researchers to date. Research into actual
logging dates and methods, perhaps using detailed lumber company records, if they exist,
would help to establish the correspondence between the human actions and the resulting
land use changes with greater precision than has been possible in this dissertation.

It is known that the stump prairie of the Kingston Plains is not a unique landscape
type in northern Michigan, but the extent-of stump prairie in Michigan has not yet been
described. In connection with that description, an overview of pre-logging forest types
and soil development would establish whether the factors contributing to stump prairie
formation in the Kingston Plains were also common in nineteenth century northern
Michigan, or whether other factors contributed to stump prairie formation at other sites.

In the Kingston Plains, pre-disturbance forest patterns are related to origin of the
stump prairie landscape. The origins of those pre-disturbance forest patterns could thus
be considered as contributors to the stump prairie landscape as well, or even an ultimate
“cause” of the stump prairie. An investigation into the Holocene site history before the

mid-18005, with a focus on forest development and forest disturbance, would provide a

164

more satisfying completion to the story of the stump prairie and also a better
understanding of the site itself.

Finally, further investigation into soil development processes following
vegetation change would help to establish whether the vegetation changes that have taken
place in the Kingston Plains have had an effect on the soil development pathways there.
Especially interesting in this regard is whether measurable degradation of the spodic
horizon can actually take place on time scales of a century or so, and whether soil
processes themselves have changed with the removal of the forest vegetation. The data
presented in this study suggest that soil degradation due to removal of forest vegetation is
minimal, or at least not easily measurable after only one century, and that soil processes
in the stump prairie remain qualitatively similar to those in the forested areas.
Nevertheless, further study is necessary to corroborate and establish these rather

preliminary conclusions.

APPENDIX A: DATA TABLES

Table 22. Particle size analysis data from the sampled pedons.

 

 

 

 

Horizon Clay Silt Sand VCS CS MS FS VFS
<0.002 0.050.002 2.0-0.05 2.0-1.0 1.0-0.5 0.05-0.25 0250.1 0.1005
mm mm mm mm mm mm mm mm
------- Wm Whaler:
EQIU
E 0.9 4.0 95.1 0.6 8.9 56.0 32.3 2.3
Bhs 4.2 4.8 91.1 0.9 7.6 52.7 36.4 2.4
le 1.7 2.0 96.3 1.1 7.0 58.4 31.6 1.9
le* 2.0 1.3 96.8 0.9 7.0 57.9 32.7 1.5
B52 0.4 0.3 99.3 2.0 12.0 62.5 22.6 0.9
BC 0.4 0.3 99.4 7.8 27.8 48.1 15.8 0.6
C 0.2 0.2 99.6 1.7 16.7 61.4 19.3 0.8
W
E 0.7 4.1 95.2 1.0 11.6 58.2 27.4 1.7
Bhs 3.3 4.6 92.1 1.9 9.2 51.9 35.1 1.8
B5] 2.0 2.3 95.7 3.3 9.6 48.9 36.9 1.2
le* 3.1 2.4 94.5 2.5 9.3 49.0 37.8 1.3
B52 0.5 0.4 99.1 0.8 10.2 69.7 18.9 0.5
BC 0.2 0.2 99.6 2.7 10.5 67.4 18.9 0.5
C 0.4 0.1 99.5 3.5 15.7 70.1 10.3 0.3
C(200)1 0.1 0.1 99.8 0.6 5.3 65.7 27.1 1.3
[308:2
E 1.4 5.1 93.6 0.5 5.5 44.0 45.7 4.3
Bhs 4.4 5.0 90.6 0.9 5.7 41.9 47.3 4.2
B5] 1.2 2.2 96.7 0.9 4.7 41.3 49.4 3.8
B5] * 2.6 2.2 95.2 0.7 4.0 32.9 57.9 4.5
B52 0.6 0.3 99.1 1.2 6.6 56.8 33.6 1.8
B52* 1.1 0.5 98.4 0.8 6.3 57.4 33.5 2.0
BC 0.2 0.1 99.6 0.4 4.4 68.0 26.4 0.9
BC" 1.0 0.3 98.7 0.2 3.9 63.8 31.0 1.1
C 0.2 0.1 99.7 0.2 4.4 71.5 23.3 0.7
C (200) 0.2 0.2 99.6 6.7 15.0 58.1 19.6 0.6
C (240) 0.2 0.1 99.7 0. 0.4 36.9 61.2 1.4

 

IIndicates depth midpoint (cm) for deep samples.

*Indicates ortstein subsample.

165

166

Table 22. (Cont.)

Horizon Clay Silt Sand VCS CS MS FS VFS
<0.002 0.05-0W 2.0-0.05 2.0-1.0 1.0-0.5 0.05-0.25 0.25-0.1 0.1-0.05

 

 

 

mm mm mm mm mm mm mm mm
...... mm ----- Msandfiaction

EOE-.4
E 0.9 4.2 94.9 1.4 14.7 58.4 24.0 1 5
Bhs 3.6 5.1 91.3 2.7 12.4 54.8 28.8 1 3
B5] 2.0 2.5 95.5 2.0 12.3 57.1 27.5 1.0
le* 3.3 2.2 94.5 2.0 10.6 55.2 31.2 1.0
B52 0.8 0.5 98.7 0.4 6.5 62.8 29.6 0 7
B52* 1.6 0.5 97.9 0.7 8.3 65.6 25.0 0.4
BC 0.3 0.2 99.5 1.7 9.0 69.3 19.5 0.5
BC“ 0.5 0.2 99.3 1.1 5.8 61.6 30.5 0.9
C 0.2 0.3 99.5 0.4 8.5 51.0 38.4 1.8
C (220) 0.1 0.1 99.8 0.2 7.9 69.7 21.7 0.5

15208::
E 1.5 5.1 93.4 1.3 15.4 63.4 18.6 1.3
B51 29 3.7 93.4 2.9 14.8 61.1 20.5 0.8
le* 2.6 1.7 95.7 2.7 13.6 61.6 21.5 0.6
B52 0.8 0.5 98.7 0.6 6.0 71.9 21.1 0.4
B52* 1.4 0.3 98.3 2.0 11.6 62.3 23.8 0.3
BC 0.3 0.3 99.4 1.0 14.4 68.2 15.8 0.6
C 0.2 0.1 99.6 0.2 10.7 77.3 11.6 0.3
C (230) 0.2 0.1 99.7 4.7 14.5 67.9 12.7 0.2

£08.15
A 9.0 7.8 83.3 0.7 10.4 55.9 30.0 2.8
E 1.4 6.2 92.4 0.9 9.5 53.4 32.4 3.7
B51 2.2 4.1 93.7 1.6 10.2 55.8 29.2 3.2
le* 2.4 3.5 94.0 1.3 9.0 54.8 31.7 3.1
B52 0.5 2.4 97.1 1.4 7.0 46.1 38.7 6.8
B52* 0.8 1.9 97.3 1.8 10.1 56.1 28.3 3.6
BC 0.2 1.3 98.5 1.6 14.9 64.8 16.7 2.0
C 0.3 1.7 98.0 5.2 16.2 48.8 27.1 2.7
2C 0.8 8.7 90.5 0.2 1.1 35.1 53.0 10.7
3C 0.1 0.5 99.5 0.3 3.7 61.6 33.9 0.5

 

*Indicates ortstein subsarnple.

167

Table 22. (Cont.)

Horizon Clay Silt Sand VCS CS MS FS VFS
<0.002 0.05-0.002 2.0-0.05 2.0-1.0 1.0-0.5 0.05-0.25 0.25-0.1 0.1-0.05

 

 

 

mm rrrm mm mm mm mm mm mm
------ Wm" Won
EQR-_7
E 0.6 4.0 95.3 0.8 11.7 62.2 24.1 1 3
Bhs 4.1 5.0 90.8 1.1 8.6 55.8 33.2 1.3
B51 2.2 1.0 96.8 0.9 10.5 64.6 23.3 0.7
le* 1.5 0.6 97.9 2.2 17.6 64.9 14.9 0.4
B52 0.4 0.2 99.4 2.0 17.3 63.8 16.4 0.4
B52* 1.0 0.4 98.6 2.0 14.8 66.2 16.7 0.3
BC 0.2 0.3 99.5 3.1 13.7 56.7 25.8 0.6
C 0.2 0.2 99.6 1.2 20.4 68.4 9.7 0.3
C (122) 2.9 2.0 95.0 4.4 22.6 57.7 14.3 0.9
C (230) 0.2 0.2 99.6 11.3 36.3 44.1 8.1 0.3
W
E 0.9 5 O 94.1 2.3 17.9 60.4 17.4 1 9
Bhsm 5.5 5.5 89.0 4.2 15.7 56.9 21.6 1.6
Bsml 2.1 2.2 95.7 3.7 19.3 60.0 15.8 1.2
Bsml 2.4 2.8 94.7 3 .0 14.9 60.7 20.1 1.3
:1:
Bsm2 0.9 0 5 98.6 5.1 24.0 55.6 14.6 0.7
2BC 0.4 0.7 98.9 21.4 41.0 29.7 7.2 0.7
3C 0.2 0.3 99.5 2.7 18.3 61.5 16.9 0.7
C (220) 0.1 0.2 99.7 2.2 10.0 71.5 16.0 0.4
£0152
E 1 2 5.1 93.8 0.8 8.0 48.2 39.6 3.4
Bhs 4 4 5.6 89.9 0.7 6.5 45.0 43.8 4.1
B51 1.6 2.7 95.7 1.0 6.8 44.0 45.2 3.0
le* 4.1 3.9 92.0 0.7 6.0 43.9 45.7 3.7
B52 0.7 1.1 98.2 0.8 5.9 50.1 41.3 1.9
B5? 1 7 1.1 97.2 0.3 3.3 51.5 42.6 2.3
BC 0.4 0.3 99.3 0.3 4.9 57.7 35.7 1.3
C 0.2 0.2 99.6 0.7 12.7 60.1 25.5 1.0
C (270) 0.4 0.5 99.2 0.2 3.0 27.9 63.3 5.6
C (280) 4.8 27. 67.4 0.1 1.0 14.3 49.2 35.4

 

*lndicates ortstein subsample.

Table 22. (Cont.)

168

 

 

 

 

Horizon Clay Silt Sand VCS CS MS FS VFS
<0.002 0.05-0112 2.0-0.05 2.0-1.0 1.0-0.5 0.05-0.25 0.25-0.l 0.1-0.05
mm mm mm mm mm mm mm mm
...... Mneearzthm Misandfiacmn
SAL-l
A 6.2 10.9 82.9 6 9 19.1 48.5 23.6 1.8
E 1.8 5.1 93.1 1 2 14.0 53. 4 29.1 2.4
le 1.2 1.8 96.9 3 4 10.9 48. 5 35.5 1.7
le* 2.0 2.1 95.8 2 5 9. 2 49. 9 36. 7 1.6
B52 0.3 0.4 99.2 2 5 21.2 58.6 17.0 0.7
BC 0.3 0.1 99.6 4 4 20.2 60.5 14.6 0.3
C 0.5 0.2 99.3 2 1 7.9 51.1 37.3 1.6
SAL-2
A 4.4 6.0 89. 7 0 7 13.0 60. 2 24.9 1.1
E 1.0 4.1 94. 9 0 9 11.9 63. 1 22.6 1.6
B51 0.7 0.8 98. 5 0 9 7.5 59.1 31.5 1.0
le* 2.0 1.4 96. 6 1 2 8.1 58. 7 31.1 0.9
B52 0.3 0.3 99. 4 6 9 23.6 57. 2 11.9 0.4
BC 0.4 0.1 99. S 5.4 27.3 59. 3 7.8 0.2
C 0.1 0.1 99. 8 0.2 5.1 81.1 13.3 0.3
M
A 3 9 8.6 87.5 0 7 6.7 51.5 38.2 2.9
E l 6 4.7 93.7 0 5 6.2 48.3 41.1 3.9
le 2.1 4.2 93.6 1 2 4.2 39.8 49. 4 5.4
le* 2.8 3.9 93. 4 1 5 3.6 37.4 51.6 5.9
B52 0.5 0.4 99.1 1 0 5.9 64.3 27. 7 1.1
BC 0 2 0.1 99. 6 l 3 6.9 70. 3 21.2 0.3
C 0 4 0.1 99.6 0 0 0.3 37. 9 60.3 1.4
SAM
A 2.9 8.6 88.5 0 6 8.2 45.9 42.6 2.6
E 0.8 5.0 94.2 0 4 6.0 41.6 48.8 3.2
le 2.0 3.2 94.8 1 0 5.6 38. 3 52.4 2.7
le* 3.2 3.5 93.2 1 3 6. 4 48. 2 42.5 1.7
B52 0.4 0.5 99. 1 2 1 13.0 57. 2 26.6 1.2
BC 0.3 0.4 99. 4 0 9 9. 8 53.7 33. 8 1.8
C 0.1 0.1 99.8 08 14.4 77.2 7.4 0.2
C (135) 0.1 0.3 99. 6 0 0 2.6 47. 7 47. 9 1.7

 

*Indicates ortstein subsample.

169

Table 22. (Cont.)

Horizon Clay 5111 Sand vcs cs MS 15 vrs
<0.002 0.050.002 2.0-0.05 2.0-1.0 1.0-0.5 0.05-0.25 025-0.] 0.1-0.05

 

 

 

mm mm mm mm mm mm mm mm
------ mm Zuzfsandfraction

SAX-.2
A 4.3 6.2 89.5 1.2 10.3 54 0 32. 4 2.1
E 1.0 4.3 94.6 1.2 8.8 56. 0 31.4 2.6
B51 1.8 3.0 95.2 2.8 9.2 53. 0 33. 0 2.0
le* 2.3 2.3 95.3 3.8 7.9 50. 6 35. 7 2.]
B52 0.6 1.1 98.3 6.5 6.4 46.2 38.5 2.3
BC 0.2 0.5 99. 3 1.5 11.4 56. 2 29. 7 1.2
C 0.1 0.1 99.8 1.8 5.7 59.7 31.2 1.6
C (250) 0.1 0.1 99. 8 1.2 9.1 69.9 19.5 0.3

SAM
A 2.3 6.7 91.0 0.4 4.6 67.9 26.1 0.9
E 1.1 3.5 95. 3 0.3 4.2 60.6 33. 6 1.3
Bhs 2.6 3.8 93. 6 0.4 3.4 59.3 35. 8 1.2
B51 1.3 2.0 96. 7 0.7 3.5 65. 8 29.3 0.7
B51* 2.3 0.9 96. 8 0.2 2.5 62. 2 34.5 0.7
B52 0.5 0.4 99.1 0.1 2.1 63. 6 33.5 0.6
B52* 1.1 0.5 98. 4 0.2 1.2 41.8 55.6 1.2
BC 0.3 0.2 99. 4 0.8 6.6 60.2 31.6 0.8
BC“ 1.0 0.6 98. 4 0.8 3.8 52.5 40.9 1.9
C 0.2 0.3 99. 5 0.0 1.1 55. 4 42. 0 1.4
C (230) 0.1 0.1 99. 8 0.3 3.6 48. 3 47.0 0.8

SAX-.Z
A 2.3 6.4 91.4 1.2 11.7 55.7 29.8 1.7
E 1.1 4.9 94.1 1.6 12. 3 54.9 29.5 1.7
B51 1.8 2.5 95.7 2.1 10. 8 54.0 32.0 1.2
le* 3.7 2.8 93.5 5.0 12. 5 47. 8 33.1 1.6
B52 0.3 0.4 99.3 0.4 6.2 59. 3 33.2 0.9
B52"' 1.3 0.5 98.2 0.3 6. 4 55. 7 36.7 1.0
BC 0.2 0.2 99.5 3.2 32. 0 58. 9 5.5 0.3
C 0.1 0.2 99.7 2.0 11.6 64. 8 21.2 0.4
c (200) 0.1 0.2 99.7 0.6 4. 5 51.4 42.4 1.2
C (300) 0.2 0 1 99.7 1.7 13. 7 70. 1 14.2 0.3

 

*Indicates ortstein subsample.

170

Table 22. (Cont.)

Horizon Clay Silt Sand VCS CS MS FS VFS
<0.002 0.05-0.002 2.0-0.05 2.0-1.0 1.0-0.5 0.05-0.25 0.25-O.l 0.1-0.05

 

 

 

mm mm mm mm mm mm mm mm
------ WWW-- 29.1mm
SAX-8

A 4.0 5.0 91 .O 0.8 11.7 56.4 29.7 1.5
E 1.2 3.9 94.9 1.3 12.2 55.6 29.3 1.7
B51 2.4 2.9 94.7 1.5 9.6 56.1 31.3 1.5
le* 2.3 1.5 96.2 2.4 11.1 55.4 30.2 0.9
B52 0.7 0.2 99.0 2.3 13.6 60.2 23.3 0.5
B52* 1.1 0.3 98.7 2.7 15.5 54.9 26.3 0.6
BC 0.3 0.3 99.5 1.5 10.7 67.0 19.9 1.0
C 0.4 0.2 99.4 9.3 38.2 43.4 8.6 0.5
C (279) 0.3 0.0 99.7 1.5 8.9 48.5 39.7 1.4

 

Table 23. Chemical data from the sampled pedons.

171

 

 

 

 

Horizon pH pH ODOE OC Fep Fed FeO AL, Ald Alo Sip SiO
2:1H202:1KC1 mi

EQR-l
0a 3.9 2.9 -- 363.8 -- -- -- -- -- -- -- --
E 4.6 3.3 0.003 2.1 0.1 0.6 0.0 0.1 0.0 0.0 0.2 0.0
Bhs 4.3 3.7 0.396 10.4 1.7 3.4 2.1 1.3 1.6 1.7 0.1 0.5
B51 4.7 4.6 0.088 3.2 0.3 1.6 0.9 1.4 1.7 3.1 0.2 1.7
le* 4.4 4.1 0.405 8.2 0.7 1.6 1.0 1.3 1.6 2.0 0.1 0.7
B52 5.2 4.8 0.030 0.9 0.1 0.4 0.2 0.6 0.4 0.8 0.1 0.6
BC 5.2 4.8 0.010 0.5 0.1 0.4 0.2 0.4 0.3 0.6 0.1 0.3
C 5.6 4.8 0.009 0.1 0.1 0.3 0.1 0.2 0.1 0.3 0.0 0.1

E01122
Oa 3.9 2.9 -- 418.1 -- -- -- -- -- -- -- --
E 4.6 3.3 0.012 4.4 0.1 0.5 0.0 0.1 0.1 0.1 0.1 0.0
Bhs 4.7 3.9 0.338 10.1 1.7 3.4 2.4 1.5 1.9 2.4 0.2 0.8
B51 4.9 4.5 0.152 4.7 0.4 2.1 1.3 1.6 2.2 3.6 0.3 2.3
le* 4.6 4.1 0.272 6.2 1.4 2.7 1.9 1.4 1.5 1.9 0.2 0.9
B52 5.] 4.6 0.060 1.4 0.3 0.8 0.4 0.8 0.8 1.2 0.1 0.7
BC 5.4 4.7 0.029 1.0 0.1 0.3 0.2 0.6 0.5 0.7 0.1 0.4
C 5.4 4.7 0.025 0.6 0.2 0.4 0.3 0.5 0.4 0.6 0.1 0.2
C(200)l 5.3 4.8 0.001 0.2 0.1 0.3 0.1 0.2 0.1 0.2 0.1 0.1

m3.
0a 3.9 2.7 -- 419.4 -- -- -- -- -- -- -- --
E 4.5 3.3 0.031 5.8 0.2 0.9 0.2 0.2 0.1 0.2 0.1 0.0
Bhs 4.6 3.8 0.593 17.3 2.0 4.3 2.8 2.0 2.7 3.6 0.2 1.8
B51 4.6 4.4 0.179 5.8 0.5 1.7 0.9 1.5 1.8 2.8 0.2 1.3
851 "' 4.5 3.9 0.599 11.4 1.3 2.2 1.5 1.6 1.7 2.1 0.2 0.4
B52 4.9 4.5 0.064 2.1 0.3 0.6 0.3 0.9 0.7 1.1 0.1 0.6
B52* 4.5 4.1 0.280 5.5 0.5 1.0 0.6 0.7 0.9 1.0 0.1 0.2
BC 5.4 4.8 0.016 0.6 0.1 0.2 0.1 0.4 0.2 0.4 0.0 0.2
BC" 4.8 4.2 0.301 5.6 0.5 0.8 0.7 0.9 0.9 1.0 0.1 0.2
C 5.5 4.7 0.014 0.5 0.1 0.2 0.1 0.2 0.1 0.3 0.0 0.1
C (200) 5.4 4.8 0.004 0.3 0.0 0.3 0.0 0.2 0.1 0.3 0.1 0.1
C(240) 5.9 4.7 0.004 0.2 0.1 0.5 0.1 0.2 0.1 0.2 0.0 0.1

 

lIndicates depth midpoint in cm for deep samples.

*Indicates ortstein subsample.

172

 

 

 

 

Table 23 (Cont.)
Horizon pH pH ODOE OC Fep Fed Feo Alp Ald Al0 Sip Sio
2:1H2021KQ ngg‘J

EQIL—A
08 4.4 3.3 -- 353.8 -- -- -- -- -- -- -- --
E 4.8 3.3 0.000 4.0 0.1 0.4 0.0 0.1 0.0 0.1 0.1 0.0
Bhs 4.5 3.7 0.392 12.6 2.0 4.2 3.2 1.1 1.7 2.3 0.1 0.7
le 4.6 4.4 0.180 6.2 0.6 2.8 1.9 1.8 2.4 4.2 0.2 2.4
B51* 4.3 4.0 0.605 10.5 1.3 2.7 2.1 1.9 2.3 3.4 0.3 1.3
B52 5.0 4.7 0.051 2.8 0.2 0.8 0.5 0.6 0.6 1.8 0.1 1.2
B52* 4.5 4.2 0.410 8.4 0.7 1.3 0.9 1.5 1.6 2.0 0.2 0.6
BC 5.4 4.8 0.020 0.4 0.1 0.3 0.2 0.4 0.3 0.7 0.1 0.4
BC* 4.9 4.6 0.063 1.7 0.2 0.5 0.3 0.6 0.6 1.1 0.1 0.4
C 5.2 4.9 0.020 0.5 0.1 0.3 0.1 0.4 0.3 0.5 0.1 0.3
C(220) 5.5 5.0 0.005 0.4 0.0 0.2 0.0 0.1 0.1 0.2 0.0 0.1

£0115
Oi 4.4 3.7 -- 282.5 -- -- -- -- -- -- -- --
E 4.5 3.2 0.038 7.6 0.2 0.6 0.2 0.2 0.1 0.1 0.3 0.0
B51 4.6 4.1 0.237 10.0 11 2.8 1.8 2.2 2.1 2.9 0.2 1.7
le* 4.6 4.0 0.380 10.6 14 2.3 1.7 2.7 2.4 2.6 0.4 0.9
B52 5.1 4.6 0.060 2.3 0.3 0.6 0.4 1.3 0.8 1.5 0.2 1.0
B52* 4.8 4.2 0.300 5.8 0.8 13 1.0 1.7 1.7 1.9 0.3 0.7
BC 5.6 4.7 0.018 1.2 0.2 0.3 0.2 0.7 0.5 0.7 0.1 0.4
C 5.5 4.8 0.017 2.9 01 0.3 0.2 0.4 0.2 0.5 0.1 0.3
C(230) 5.7 4.8 0.009 0.3 01 0.2 0.1 0.3 0.1 0.3 0.1 0.1

£0815
Oi 4.1 3.5 - 372.3 -- -- -- -- -- -- -- --
A 4.9 3.4 0.083 57.1 0.6 2.0 0.9 0.6 1.0 0.9 0.9 0.0
E 4.6 3.5 0.028 6.7 0.3 0.8 0.2 0.2 0.1 0.2 0.2 0.0
B51 5.] 4.4 0.230 7.8 0.7 3.3 2.2 2.0 2.2 4.4 0.2 2.5
le* 5.0 4.3 0.322 8.6 0.7 2.2 1.2 1.5 2.3 2.8 0.2 1.4
B52 5.2 4.6 0.050 2.0 0.2 0.8 0.4 0.8 0.6 1.1 0.1 0.7
B52* 5.2 4.6 0.108 3.0 03 0.9 0.6 1.2 1.0 1.6 0.2 0.9
BC 5.3 4.7 0.015 0.7 01 0.2 0.1 0.4 0.2 0.6 0.1 0.4
C 5.4 4.9 0.010 0.5 01 0.3 0.1 0.4 0.2 0.4 0.1 0.3
2C 5.0 4.3 0.050 3.8 03 0.8 0.3 0.5 0.5 0.6 0.1 0.2
3C 5.6 5.0 0.002 0.1 01 0.2 0.1 0.2 0.1 0.2 0.1 0.1

 

*Indicates ortstein subsample.

Table 23 (Cont.)

173

 

 

 

 

Horizon pH pH ODOE OC Fep Fed Feo Alp A1,, A10 Sip Sio
2:1 11,021Kc1 W

EQR-Z
0a 4.7 4.3 -- 215.6 -- -- -- -- -- -- -- --
E 4.6 3.7 0.000 3.3 0.1 0.3 0.0 0.1 0.0 0.0 0.1 0.0
Bhs 4.7 3.7 0.545 16.9 2.6 51 2.9 1.5 1.9 1.8 0.1 0.5
B51 4.8 4.5 0.190 5.5 0.4 1 9 1.5 2.2 1.8 7.1 0.2 4.5
le* 5.1 4.3 0.302 6.9 0.7 1.6 1.2 2.1 1.9 2.5 0.2 1.6
B52 5.5 4.9 0.030 1.2 0.2 0.4 0.3 0.8 0.5 1.3 0.2 0.8
Bs2* 5.3 4.6 0.180 4.5 0.4 0.7 0.6 1.5 1.3 2.6 0.2 1.3
BC 6.0 4.9 0.022 1.2 O l 0.3 0.2 0.5 0.4 0.5 0.1 0.3
C 6.1 5.0 0.000 0.4 0.1 0.1 0.1 0.2 0.1 0.2 0.1 0.1
C (122) 5.6 4.6 0.030 1.1 0.3 - 0.9 0.3 0.8 0.4 0.9 0.3 0.4
C (230) 5.7 5.0 0.010 0.3 0 1 0.3 0.1 0.2 0.1 0.3 0.1 0.2

EQM
0a 4.2 3.2 -- 318.3 -- -- -- -- -- -- -- --
E 4.4 3.2 0.010 4.1 0 1 0.4 0.1 0.1 0.0 0.1 0.1 0.0
Bhsm 4.5 3.8 0.621 20.9 3.1 5.5 3.3 2.7 2.7 2.7 0.5 0.7
Bsml 4.8 4.4 0.199 5.7 0.5 2.0 1.2 1.7 2.0 3.3 0.2 2.1
Bsml* 5.1 4.4 0.380 10.1 0.9 2.5 1.6 3.1 2.9 4.3 0.3 2.2
Bsm2 5.3 4.6 0.067 2.1 0.2 0.6 0.4 1.2 0.6 1.6 0.2 0.9
2BC 5.4 4.8 0.020 1.1 0 1 0.5 0.2 0.7 0.3 0.9 0.1 0.5
3C 5.8 4.8 0.017 0.6 0 l 0.2 0.1 0.3 0.2 0.4 0.1 0.2
C (220) 5.9 4.8 0.000 0.0 0 1 0.2 0.1 0.2 0.0 0.1 0.1 0.0

EQR:2
Oi 3.8 3.1 -- 192.7 -- -- -- -- -- -- -- -
E 4.4 3.2 0.005 4.7 0.1 0.8 0.1 0.1 0.1 0.2 0.2 0.0
Bhs 4.4 3.6 0.481 13.9 2.0 3.8 2.1 1.6 1.7 1.6 0.2 0.4
le 4.7 4.5 0.160 6.0 0.5 1.9 1.1 2.1 2.0 4.1 0.3 2.6
351* 4.6 3.9 0.560 13.6 1.4 3.1 2.3 1.8 2.1 3.5 0.2 1.2
B52 5.1 4.6 0.038 2.5 0.2 0.8 0.3 0.6 0.7 1.0 0.1 0.6
B52* 5.0 4.2 0.440 8.7 0.8 1.5 0.9 2.5 2.1 2.4 0.2 0.6
BC 5.4 4.8 0.021 0.7 0.2 0.4 0.1 0.5 0.4 0.6 0.1 0.3
C 5.6 4.9 0.010 0.6 01 0.3 0.1 0.3 0.3 0.4 0.1 0.2
C (270) 5.5 -- 0.018 0.3 0.1 0.5 0.2 0.2 0.2 0.4 0.0 0.2
C (280) 5.6 4.3 0.031 0.5 0.0 4.8 0.6 0.8 1.3 1.4 0.1 0.6

 

*Indicates ortstein subsample.

174

 

 

 

 

Table 23 (Cont.)
Horizon pH pH ODOE OC Fep Fed Feo Alp Ald Alo Sip Si0
2:1 11202-1Kc1 mi
SAK—l
A 4.4 3.3 -- 96.0 -- -- -- -- -- -- -- --
E 4.7 3.5 0.050 10.1 0.4 0.8 0.4 0.4 0.2 0.3 0.2 0.0
B5] 5.3 4.7 0.060 1.8 0.3 1 l 0.6 1.3 1.1 2.0 0.3 1.2
le* 5.2 4.4 0.272 6.3 0.9 1.8 1.3 2.1 1.9 2.6 0.3 1.1
B52 5.5 4.9 0.010 0.5 0 1 0.4 0.3 0.4 0.3 1.1 0.1 0.6
BC 5.6 5.0 0.000 0.3 0.1 0.2 0.1 0.3 0.1 0.5 0.0 0.3
C 5.6 4.6 0.020 0.1 0.3 0.5 0.2 0.5 0.3 0.3 0.2 0.]
SA V-Z
A 4.4 3.4 -- 40.0 -- -- -- -- -- -- -- --
E 4.7 3.6 0.015 5.2 0.1 0.4 0.1 0.2 0.1 0.1 0.2 0.0
le 5.5 4.9 0.030 1.3 0.2 1 0 0.6 0.9 0.7 2.3 0.2 1.5
B51* 5.1 4.4 0.392 6.8 0.8 1.9 1.4 2.3 2.0 3.1 0.1 1.8
B52 5.6 5.0 0.010 0.4 01 0.2 0.2 0.4 0.2 0.7 0.1 0.5
BC 5.8 4.9 0.005 0.3 0 1 0.4 0.2 0.3 0.2 0.4 0.1 0.2
C 6.0 4.9 0.012 0.1 0 l 0.2 0.1 0.2 0.0 0.2 0.0 0.2
SAL/:3
A 4. 1 3 .3 -- 52. 1 -- -- -- -- -- -- -- --
E 4.2 3.5 0.018 5.9 0.2 0.6 0.2 0.2 0.1 0.2 0.2 0.0
B5] 4.9 4.6 0.171 7.6 0.6 2.8 1.3 2.2 2.5 4.0 0.4 2.6
le" 5.0 4.4 0.430 11.4 1 3 4.1 2.8 4.3 4.0 5.7 0.2 2.7
B52 5.3 4.8 0.022 1.2 01 0.4 0.2 0.6 0.4 0.8 0.1 0.5
BC 5.7 4.9 0.012 0.6 0.1 0.3 0.1 0.4 0.3 0.5 0.1 0.3
C 5.8 4.9 0.000 0.1 0.0 0.2 0.1 0.2 0.1 0.4 0.0 0.3
SAM
A 4.6 3.4 -- 42.6 -- -- -- -- -- -- -- --
E 4.5 3.7 0.012 5.0 0.0 0.7 0.1 0.0 0.1 0.1 0.0 0.0
B5] 4.6 4.6 0.153 5.1 0.5 2.7 1.3 1.6 1.8 4.6 0.3 3.2
le* 5.3 4.3 0.485 11.2 1 3 3.6 1.9 2.6 3.5 4.3 0.2 2.0
B52 4.6 4.9 0.023 1.3 0.1 0.7 0.4 0.5 0.4 1.4 0.1 1.0
BC 5.4 4.8 0.020 0.7 0.2 0.7 0.3 0.5 0.4 0.7 0.1 0.4
C 5.3 4.9 0.000 0.4 01 0.1 0.1 0.3 0.1 0.3 0.1 0.2
C (135) 5.3 4.9 0.005 0.4 0 1 0.3 0.1 0.3 0.2 0.4 0.1 0.2

 

*Indicates ortstein subsample.

175

 

 

 

 

Table 23 (Cont.)
Horizon pH pH ODOE OC Fep Fed Feo Alp A1,, A10 Sip Sio
2:1 H2021KCI and

SA [.5
A 4.5 3 .0 -- 5 5 .0 -- -- -- -- -- -- -- --
E 4.6 3.5 0.018 4.5 0.2 0.7 0.1 0.2 0.1 0.2 0.2 0.0
le 5.3 4.6 0.142 5.2 0.6 2.9 1.4 1.8 2.1 3.9 0.3 2.4
le* 4.6 4.4 0.418 9.8 1.2 3.5 2.2 2.8 3.7 4.3 0.2 1.7
B52 5.4 4.9 0.020 0.8 0.2 1.0 0.4 0.7 0.5 1.6 0.2 1.1
BC 5.3 5.0 0.002 0.4 0 l 0.3 0.1 0.4 0.2 0.7 0.1 0.4
C 5.3 4.9 0.002 0.3 0.1 0.3 0.1 0.3 0.1 0.4 0.1 0.2
C (250) 5.7 5.0 0.001 -0.3 0.0 0.2 0.0 0.1 0.0 0.2 0.0 0.1

SAM
A 4.2 3.3 -- 39.5 -- -- -- -- -- -- -- --
E 4.8 3.5 0.015 4.5 0 l 0.4 0.1 0.1 0.0 0.1 0.1 0.1
Bhs 4.8 3.9 0.237 7.3 1.1 2.1 1.1 0.7 0.7 0.9 0.1 0.3
B5] 5.0 4.3 0.161 5.1 0.5 2.0 1.2 1.4 1.7 3.5 0.1 2.4
le* 5.0 4.1 0.438 8.0 0.9 1.9 1.1 1.3 1.4 1.7 0.3 0.6
B52 5.2 4.6 0.098 1.5 0.1 0.7 0.5 0.6 0.5 2.3 0.1 2.1
B52* 5.2 4.4 0.250 5.7 0.8 1.7 1.1 1.7 2.1 2.6 0.2 1.1
BC 5.4 4.8 0.013 0.8 0.1 0.3 0.2 0.5 0.4 0.9 0.1 0.7
BC“ 5.4 4.4 0.239 5.1 0.5 1.1 0.6 1.3 1.5 1.6 0.2 0.5
C 5.6 4.9 0.004 0.7 0.1 0.3 0.1 0.3 0.2 0.6 0.1 0.4
C (230) 5.7 4.9 0.000 0.3 0.0 0.2 0.0 0.2 0.1 0.3 0.0 0.2

SAEZ
A 4. 1 3 .4 -- 33. 1 -- -- -- -- -- -- -- --
E 4.7 3.5 0.016 4.5 0.1 0.5 0.1 0.1 0.0 0.2 0.2 0.]
B51 5.1 4.4 0.163 5.5 0.4 2.1 1.3 1.4 1.8 5.1 0.2 3.5
le * 5.1 4.2 0.520 11.9 1.0 3.0 1.9 1.9 2.7 3.8 0.1 1.3
B52 5.3 4.7 0.041 1.8 0.2 0.7 0.3 0.7 0.6 1.6 0.1 1.0
B52* 5.2 4.3 0.250 4.8 0.5 1.1 0.6 1.1 1.3 1.8 0.1 0.7
BC 5.7 4.8 0.023 0.6 0 l 0.3 0.1 0.4 0.3 0.4 0.1 0.3
C 5.7 4.9 0.009 0.5 0.1 0.3 0.1 0.3 0.2 0.4 0.1 0.3
C (200) 5.9 4.9 0.010 0.3 0.0 0.3 0.0 0.2 0.1 0.3 0.1 0.2
C (300) 5.8 4.9 0.000 0.1 0.0 0.2 0.1 0.1 0.0 0.3 0.0 0.2

 

*Indicates ortstein subsample.

Table 23 ( Cont.)

176

 

 

 

 

Horizon pH pH ODOE OC Fep Fed Feo Alp A1,, A10 Sip Sio
2:1H2021KC1 gkgi
SAL-S
A 4.4 3.3 -- 29.0 -- -- -- -- -- -- -- --
E 5.1 3.8 0.020 3.5 0.2 0.5 0.1 0.2 0.1 0.2 0.1 0.]
B51 5.8 4.7 0.143 4.0 0.5 2.4 1.4 1.4 1.6 3.6 0.3 2.5
le* 6.1 4.7 0.362 7.5 0.6 2.3 1.1 1.5 2.6 3.2 0.2 1.5
B52 6.1 5.1 0.032 1.4 01 0.7 0.3 0.4 0.4 1.3 0.1 0.9
B52* 6.0 4.9 0.150 3.7 03 0.7 0.4 1.3 1.1 1.8 0.2 0.9
BC 6.1 5.1 0.020 0.7 01 0.3 0.1 0.3 0.3 0.7 0.1 0.4
C 6.3 5.2 0.101 0.2 0.1 0.3 0.1 0.2 0.1 0.3 0.0 0.2
C(279) 5.9 4.9 0.000 0.6 0.0 03 0.0 0.1 0.0 0.2 0.0 0.1

 

*Indicates ortstein subsample.

Table 24. Fe and Al ratio data from the sampled pedons.

177

 

 

(Feo-Fep)/ (Alo-Alpy (A1,-A1,)/
Horizon Fep FeO/Fed Alp Ala/Ald Alo-Alp Sio Fey/Alp
W

EQRJ
E -0.67 0.05 -059 0.83 0.0 3.39 0.46
Bhs 0.28 0.63 0.35 1.09 0.5 0.92 0.63
B51 1.61 0.57 1.25 1.77 1.7 1.07 0.12
le* 0.48 0.64 0.53 1.26 0.7 1.00 0.25
852 0.40 0.54 0.44 1.98 0.3 0.46 0.12
BC 0.51 0.43 0.29 1.73 0.1 0.40 0.12
C 0.25 0.27 0.31 2.28 0.1 0.67 0.13

EQ&2
E -033 0.10 -031 1.14 0.0 2.06 0.41
Bhs 0.45 0.72 0.53 1.25 0.8 1.07 0.53
B51 2.04 0.62 1.28 1.68 2.0 0.90 0.13
le* 0.42 0.71 0.40 1.24 0.6 0.66 0.48
852 0.63 0.50 0.56 1.53 0.4 0.63 0.16
BC 0.21 0.52 0.13 1.60 0.1 0.25 0.11
C 0.25 0.67 0.14 1.54 0.1 0.28 0.20
C (200)1 -001 0.22 0.00 1.48 0.0 0.00 0.14

508:1
E -025 0.17 0.00 1.49 0.0 0.02 0.45
Bhs 0.41 0.66 0.79 1.33 1.6 0.93 0.47
B5] 0.74 0.52 0.87 1.59 1.3 1.03 0.16
le"' 0.23 0.72 0.35 1.26 0.6 1.58 0.39
B52 0.12 0.56 0.30 1.64 0.3 0.48 0.16
B5? 0.31 0.65 0.46 1.21 0.3 1.51 0.34
BC .010 0.39 0.18 2.03 0.1 0.41 0.14
BC“ 0.32 0.86 0.14 1.06 0.1 0.84 0.30
C -004 0.32 0.28 2.48 0.1 0.56 0.13
C (200) 0.02 0.15 0.42 2.37 0.1 0.81 0.09
C (240) 0.54 0.22 0.16 2.44 0.0 0.40 0.23

 

lIndicates depth midpoint in cm for deep samples.
*Indicates ortstein subsample.

-4‘-

178

Table 24. (Cont.)

 

 

(Fee-Fep)/ (Alo-Alp)/ (Alo-Alp)/
Horizon Fep Feo/Fed Alp Alo/Ald Alo-Alp Sio Fgfiklp
238:1

EQR-A
E -0.67 0.08 -0.19 3.33 0.0 -1.15 0.43
Bhs 0.63 0.76 1.06 1.38 1.2 1.88 0.84
B51 2.41 0.68 1.36 1.75 2.4 1.03 0.15
351* 0.61 0.78 0.78 1.51 1.5 1.18 0.32
B52 2.10 0.65 2.24 3.00 1.3 1.11 0.14
B52* 0.25 0.69 0.32 1.24 0.5 0.87 0.22
BC 0.34 0.51 0.64 1.99 0.3 0.75 0.14
BC“ 0.71 0.62 0.95 1.84 0.5 1.30 0.15
C -0.03 0.41 0.15 1.86 0.1 0.26 0.15
C (220) 0.09 0.16 0.33 2.17 0.0 0.41 0.12

E0322
E -0.07 0.33 -0.21 1.89 0.0 -1.08 0.67
B51 0.62 0.66 0.34 1.38 0.7 0.44 0.25
le * 0.25 0.76 -0.04 1.06 -0.1 -0.12 0.25
B52 0.43 0.64 0.22 2.01 0.3 0.29 0.11
B52* 0.25 0.76 0.12 1.10 0.2 0.31 0.22
BC 0.15 0.60 0.06 1.57 0.0 0.11 0.12
C 0.84 0.59 0.13 2.18 0.1 0.23 0.12
C (230) 0.12 0.49 -0.07 2.62 0.0 -0.17 0.13

EQflzé
A 0.48 0.45 0.45 0.89 0.3 ~108.47 0.49
E -0.23 0.31 -0.30 1.30 -0.1 -3.85 0.69
B5] 2.37 0.66 1.19 1.95 2.4 1.00 0.16
B51 * 0.83 0.55 0.81 1.21 1.2 0.95 0.21
B52 1.48 0.52 0.51 2.07 0.4 0.59 0.11
B52* 1.22 0.63 0.39 1.63 0.5 0.55 0.11
BC 0.44 0.67 0.74 3.43 0.3 0.71 0.14
C 0.24 0.53 -0.05 2.68 0.0 -0.08 0.12
2C 0.19 0.37 0.18 1.07 0.1 0.46 0.25
3C 0.02 0.35 0.14 4.42 0.0 0.32 0.14

 

*lndicates ortstein subsample.

Table 24. (Cont.)

179

 

 

(F eo-F ep)/ (A1,-A1,)/ (Alo-Alp)/
Horizon Fep Fee/Fed Alp Alo/Ald Alo-Alp Sio Fep/Alp
15123-7
E -0.67 0.09 -0.48 -1.09 0.0 1.33 0.61
Bhs 0.13 0.57 0.17 0.93 0.3 0.51 0.83
B51 3.23 0.81 2.18 3.88 4.9 1.12 0.08
B51* 0.70 0.74 0.17 1.33 0.4 0.23 0.16
B52 0.61 0.59 0.51 2.61 0.4 0.54 0.09
B52* 0.60 0.83 0.74 2.00 1.1 0.87 0.13
BC 0.19 0.53 0.13 1.31 0.1 0.22 0.13
C -0.11 0.42 —0.01 3.19 0.0 -0.02 0.13
C (122) 0.21 0.34 0.06 2.20 0.1 0.12 0.15
C (230) 0.62 0.46 0.37 4.14 0.1 0.45 0.21
£01233
E -0.36 0.21 -0.21 2.60 0.0 1.79 0.56
Bhsm 0.05 0.59 0.01 1.03 0.0 0.04 0.56
Bsml 1.57 0.60 0.91 1.70 1.6 0.80 0.13
Bsml * 0.82 0.65 0.41 1.47 1.3 0.60 0.14
Bsm2 0.90 0.62 0.37 2.47 0.4 0.49 0.08
2BC 0.46 0.43 0.31 2.84 0.2 0.41 0.10
3C 0.21 0.60 0.23 1.88 0.1 0.35 0.16
C (220) 0.10 0.31 -0.31 5.45 -0.1 -2.27 0.17
EQR:2
E -0.18 0.14 0.35 2.86 0.0 1.00 0.55
Bhs 0.03 0.55 0.01 0.93 0.0 0.04 0.63
B5] 1.37 0.61 0.93 2.02 2.0 0.80 0.11
le* 0.58 0.74 0.96 1.63 1.7 1.53 0.39
B52 0.67 0.34 0.67 1.47 0.4 0.66 0.13
B52* 0.18 0.62 -0.03 1.18 -0.1 -0.14 0.15
BC -0.10 0.32 0.15 1.49 0.1 0.26 0.14
C -0.19 0.26 0.22 1.51 0.1 0.35 0.17
C (270) 2.56 0.39 0.98 1.98 0.2 1.01 0.12
C (280) 10.89 0.12 0.73 1.08 0.6 1.01 0.03

 

*Indicates ortstein subsample.

180

Table 24. (Cont.)

 

 

(F eo-F ep)/ (AID-Alpy (A1,-A1,)/
Horizon Fep Feo/Fed Alp AlO/Ald Alo-Alp Sio Pep/Alp
eke”
SAL/J
E 0.01 0.48 -O.17 1.83 -0.1 -1.92 0.50
B5] 0.74 0.50 0.56 1.81 0.7 0.61 0.12
B51 * 0.52 0.71 0.27 1.34 0.6 0.51 0.20
B52 1.32 0.74 1.46 3.94 0.7 1.07 0.16
BC 0.27 0.50 0.71 5.50 0.2 0.77 0.16
C -0.20 0.48 -0.39 1.28 -0.2 -1.57 0.27
SAM
E -0.35 0.17 -0.30 2.46 -0.1 -1 .19 0.30
B5] 2.35 0.57 1.45 3.06 1.3 0.91 0.09
B51 * 0.78 0.76 0.33 1.55 0.8 0.45 0.17
B52 0.81 0.79 0.70 4.01 0.3 0.60 0.12
BC 0.21 0.44 0.26 1.77 0.1 0.39 0.22
C 0.53 0.53 0.24 6.19 0.0 0.26 0.15
SAIL-3
E 0.02 0.26 0.09 2.38 0.0 0.42 0.39
B51 1.04 0.45 0.88 1.61 1.9 0.75 0.14
le * 1.12 0.67 0.33 1.42 1.4 0.54 0.15
B52 0.30 0.43 0.28 1.87 0.2 0.36 0.11
BC 0.12 0.40 0.17 1.91 0.1 0.30 0.10
C 0.80 0.27 1.38 7.92 0.3 1.05 0.09
SAM
E 0.00 0.12 0.00 2.28 0.1 3.82 0.00
B5] 1.85 0.50 1.89 2.58 3.0 1.00 0.14
B51 * 0.44 0.52 0.69 1.23 1.8 0.90 0.25
B52 2.64 0.51 1.79 3.25 0.9 0.95 0.10
BC 0.46 0.36 0.46 1.82 0.2 0.60 0.18
C -0.25 0.48 0.13 3.21 0.0 0.23 0.17
C (135) -0.06 0.26 0.34 2.45 0.1 0.48 0.13

 

*Indicates ortstein subsample.

 

Table 24. (Cont.)

181

 

 

(Fee-Fepy (A1,-A1,)/ (Alo-Alpy
Horizon Fep Fee/Fed Alp Alo/Ald AlO-Alp Sio Pep/Alp
and
54 V-5
E -0.28 0.19 0.00 2.54 0.0 -0.03 0.38
851 1.28 0.46 1.13 1.85 2.0 0.89 0.16
le* 0.84 0.62 0.51 1.16 1.4 0.90 0.20
B52 1.23 0.35 1.49 3.57 1.0 0.94 0.12
BC 0.80 0.40 0.76 3.59 0.3 0.67 0.09
C 0.02 0.34 0.48 3.65 0.1 0.56 0.17
C (250) 1.37 0.24 0.61 6.23 0.1 0.58 0.07
SAM
E -0.31 0.17 0.04 6.05 0.0 0.12 0.35
Bhs 0.04 0.52 0.26 1.18 0.2 0.55 0.75
B5] 1.29 0.60 1.57 2.07 2.1 0.95 0.19
le" 0.19 0.60 0.30 1.24 0.4 0.70 0.34
B52 3.01 0.67 2.74 4.57 1.7 0.80 0.09
B52* 0.45 0.63 0.48 1.24 0.8 0.83 0.21
BC 0.54 0.50 0.75 2.52 0.4 0.58 0.10
B0“ 0.20 0.54 0.26 1.10 0.3 0.62 0.18
C 0.03 0.28 0.67 2.97 0.2 0.53 0.10
C(230) -0.11 0.18 0.14 3.12 0.0 0.16 0.10
SALZ
E -0.33 0.18 0.40 3.27 0.0 0.55 0.62
B51 2.44 0.64 2.58 2.82 3.7 1.10 0.13
le" 0.86 0.64 1.01 1.42 1.9 1.54 0.27
B52 0.82 0.45 1.46 2.82 1.0 1.00 0.13
B5?“ 0.37 0.56 0.58 1.40 0.6 0.99 0.20
BC -0.03 0.36 0.09 1.45 0.0 0.12 0.13
C -0.13 0.32 0.18 1.93 0.1 0.18 0.15
C (200) 0.13 0.14 0.92 3.05 0.2 0.71 0.11
C (300) 3.48 0.37 1.26 6.37 0.1 0.73 0.07
SAM
E -0.57 0.19 -0.29 1.71 -0.1 -l.22 0.44
851 1.96 0.58 1.62 2.25 2.2 0.94 0.16
le* 0.96 0.50 1.11 1.23 1.7 1.12 0.18
B52 1.85 0.40 1.96 3.05 0.9 1.00 0.10
B52* 0.33 0.65 0.45 1.60 0.6 0.67 0.13
BC 0.71 0.39 1.26 2.21 0.4 0.97 0.10
C 0.77 0.31 0.92 3.37 0.2 0.83 0.14
C (279) 1.29 0.15 0.53 4.75 0.1 0.88 0.07

 

*Indicates ortstein subsample.

182

Table 25. Cations sorbed on cation exchange resins in the sampled pedons.

 

 

 

 

Pedon Set Horizon Mg Fe Ca Al
WW

FOR-5 1 E 25 21 58 89
2 E 38 13 1 12 15
3 E 50 23 1758 39
4 E 36 12 172 9
l B 14 17 18 47
2 B 45 69 282 144
4 B 28 58 81 206
1 BC 173 14 708 100
2 BC 14 10 46 22
3 BC 28 24 67 71
4 BC 17 10 78 15

FOR-7 1 E 178 13 503 8
2 E 252 17 988 16
3 E 280 27 941 76
4 E 36 12 72 12
1 B 1 19 29 385 67
2 B 75 66 168 440
3 B 135 54 315 146
4 B 140 233 256 660
1 BC 47 13 95 65
2 BC 33 12 88 85
3 BC 10 10 20 48
4 BC 53 12 94 83

FOR-9 1 E 39 15 105 56
2 E 28 14 25 49
3 E 55 25 106 19
4 E 38 15 132 25
1 B 16 25 20 40
2 B 59 41 79 90
3 B 51 26 69 51
4 B 37 51 36 326
1 BC 4 l 1 -1 l8
2 BC 5 15 -3 21
3 BC 14 17 7 61
4 BC 20 32 37 70

 

 

183

Table 25. (Cont.)

 

 

 

 

Pedon Set Horizon Mg Fe Ca Al
Wm

SAV-6 1 E 25 l 1 25 10
2 E 50 18 84 27
3 E 43 13 80 9
4 E 53 12 67 6
1 B 23 31 292 53
2 B 29 28 68 77
3 B 37 16 94 18
4 B 60 26 151 44
1 BC 6 l l 18 27
2 BC 15 21 23 29
3 BC 25 19 61 66
4 BC 30 14 103 81

SAV-7 1 E 81 13 157 8
2 E 47 12 92 14
3 E 33 1 l 109 12
4 E 35 1 l 54 8
1 B 34 26 89 64
2 B 66 42 244 97
3 B 31 47 256 1 17
4 B 51 26 182 129
1 BC 8 l 1 24 22
2 BC 26 14 72 57
3 BC 27 17 69 73

SAV-8 1 E 17 12 41 9
2 E 20 l l 61 7
3 E 24 13 58 10
4 E 6 15 14 8
1 B 43 19 125 26
2 B 34 21 105 21
3 B 7 33 23 53
4 B 29 16 78 16
1 BC 17 17 40 27
2 BC 1 l 12 36 7
3 BC 16 26 28 50
4 BC 10 12 37 25

184

Table 26. Cations sorbed on chelating resins in the sampled pedons.

 

 

Pedon Set Horizon Fe Al
----- W m-----
FOR-5 1 E 89 229
2 E 164 410
3 E 89 216
4 E 125 389
l B 199 959
4 B 42 107
1 BC 34 796
4 BC 29 154
FOR-7 1 E 62 130
2 E 37 127
3 E 145 347
4 E 108 400
1 B 3 l 1 876
2 B 176 1677
3 B 349 2039
4 B 1005 2228
1 BC 14 152
2 BC 15 109
3 BC 49 374
4 BC 31 258
FOR-9 1 E 85 571
2 E 46 200
3 E 15 66
4 E 29 177
l B 83 252
2 B 265 753
3 B 56 395
4 B 69 304
1 BC 0 106
2 BC 7 107
3 BC 19 137
4 BC 7 105

 

185

Table 26. (Cont.)

 

 

Pedon Set Horizon Fe Al
----- WWW
SAV-6 1 E 36 129
2 E 34 130
3 E 40 l 10
4 E 26 152
1 B 159 1 153
2 B 103 735
3 B 147 434
4 B 242 596
1 BC 6 130
2 BC 6 69
3 BC 12 178
4 BC 18 147
SAV-7 l E 137 215
2 E 27 126
3 E 47 172
4 E 26 75
1 B l 16 1010
2 B 165 715
3 B 88 331
4 B 125 540
1 BC 16 178
2 BC 21 201
3 BC 38 193
4 BC 27 275
SAV-8 1 E 44 238
2 E 22 147
3 E 10 80
4 E 2 69
l B 97 314
2 B 49 133
3 B 190 454
4 B 91 194
1 BC 10 88
2 BC 13 65
3 BC 14 84
4 BC 105 320

 

186

APPENDIX B: PEDON DESCRIPTIONS

Pedon: FOR-1

Location: SW 1/4 of SW 1/4 of NE 1/4 of section 19, T 48 N, R 15 W
Date of description: July 12, 1994

Series: Kalkaska

Subgroup classification: Typic Haplorthod

Parent material: Outwash sands

Physiography: Outwash plain

Slope: < 2 %

Comments: Forested area with mature trees, marked for logging.

Oa 0-3 cm. Very dark gray (10 YR 3/1) well decomposed hardwood leaf litter;
abrupt smooth boundary.

E 3-18 cm. Grayish brown (10 YR 5/2) sand; single grain; loose; many medium
and fine roots; abrupt wavy boundary.

Bhs 18-23 cm. Very dark gray (7.5 YR 3/2) sand; single grain; loose; many medium
and fine roots; very dark gray (7.5 YR 3/2) weakly cemented ortstein
occurs in chunks less than 1 cm diameter and occupies about 20 percent
of exposed surface; clear wavy boundary.

B51 23-48 cm. Strong brown (7.5 YR 4/6) sand; single grain; loose; few fine roots;
dark reddish brown (5 YR 3/4) very strongly cemented ortstein occurs in
columns 10 cm wide and 50 cm long and occupies about 35 percent of
exposed surface; gradual wavy boundary.

B52 48-78 cm. Yellowish brown (10 YR 5/6) slightly gravelly sand; single grain;

loose; gradual wavy boundary.
BC 78-103 cm. Yellowish brown (10 YR 5/4) sand; single grain; loose; gradual
smooth boundary.
C 103-150 cm. Yellowish brown (10 YR 6/4) sand; single grain; loose.
Pedon: FOR-2

Location: NW 1/4 of NE 1/4 of NW 1/4 of section 27, T 48 N R 15 W.
Date of description: July 24, 1994

Series: Deer Park

Subgroup classification: Spodic Udipsamment

Parent material: Outwash sand

Physiography: Outwash plain

Slope: < 2 %

Comments: Evidence of logging about 10 - 20 years ago.

08 0 - 3 cm. Black (N 2/0) well decomposed hardwood leaf litter; many fine roots.

E 3 - 18 cm. Grayish brown (10 YR 5/2) sand; weak fine subangular blocky; very
friable; many fine and medium roots; clear wavy boundary.

187

Bhs 18 - 23 cm. Dark reddish brown (5 YR 2.5/2) sand; weak fine subangular
blocky; friable; many fine and medium roots; dark reddish brown (5 YR
2.5/2) strongly cemented ortstein occurs in columns 14 cm wide and up
to 70 cm long and occupies about 25 percent of exposed surface; clear
irregular boundary.

B51 23 - 43 cm. Strong brown (7.5 YR 4/6) sand; weak medium subangular blocky;
friable; common fine roots; dark reddish brown (5 YR 3/2) strongly
cemented ortstein occurs in columns 20 cm wide and 70 cm long and
occupies about 20 percent of exposed surface; gradual wavy boundary.

B52 43 - 78 cm. Dard yellowish brown (10 YR 4/6) sand; weak medium subangular
blocky; very friable; few fine roots; dark brown (7.5 YR 4/4) medium
cemented ortstein occurs in columns 5 to 10 cm wide and occupies about
15 percent of exposed surface; gradual wavy boundary.

BC 78 - 103 cm. Yellowish brown (10 YR 5/4) sand; weak medium subangular
blocky; very friable; strong brown (7.5 YR 4/6) medium cemented
ortstein occurs in columns 5 cm wide and occupies about 5 percent of
exposed surface; gradual wavy boundary.

C 103 - 150 cm. Yellowish brown (10 YR 5/4) sand; massive.
C 190 - 210 cm.
Pedon: FOR-3

Location: NE 1/4 of SW 1/4, section 1 1, T 48 N, R 15 W

Date of description: July 25, 1994

Series: Kalkaska

Subgroup classification: Typic Haplorthod

Parent material: Outwash sand

Physiography: Outwash plain

Slope: < 2 %

Comments: Some logging has been done in the last 10-20 years.

08 0 - 4 cm. (7.5 YR 3/2) well-decomposed leaf litter.

E 4 - 16 cm. Grayish brown (10 YR 5/2) sand; weak medium subangular blocky
structure; very firable; common medium roots; abrupt wavy boundary.

Bhs 16 - 19 cm. Dark brown (7.5 YR 3/2) sand; moderate medium subangular
blocky structure; friable; common medium roots; black (5 YR 2.5/1)
strongly cemented ortstein occurs in columns 15 - 20 cm wide which
extend into the B51 horizon and occupies about 40 percent of the exposed
surface; abrupt wavy boundary.

B51 19 - 39 cm. Dark yellowish brown (7.5 YR 4/6) sand; weak medium subangular
blocky structure; very friable; common fine roots; black (5 YR 2.5/ 1)
very strongly cemented orstein occurs in columns and chunks 15 - 20 cm
wide which extend into lower horizons and occupies about 40 percent of

the exposed surface; gradual wavy boundary.

B52 39 - 74 cm. Dark yellowish brown (10 YR 4/6) sand; weak coarse subangular
blocky structure; very friable; few fine roots; dark brown (7.5 YR 3/4)
very strongly cemented ortstein occurs in columns and chunks 15 - 20 cm

 

188

wide which exted into lower horizons and occupies about 30 percent of
the exposed surface; gradual wavy boundary.

BC 74 - 114 cm. Yellowish brown (10 YR 5/4) sand; massive; dark brown (7.5 YR
3/4) strongly cemented ortstein occurs in columns about 10 cm wide and
occupies about 10 percent of the exposed surface; gradual wavy
boundary.

C 114 - 150 cm. Light yellowish brown (10 YR 6/4) sand; massive.

C 190 - 210 cm.

C 230 - 250 cm.

Pedon: FOR-4

Location: NE 1/4 of SE 1/4 of SW 1/4 of SE 1/4, section 8, T 48 N, R 15 W.
Date of description: July 29, 1994

Series: Kalkaska

Subgroup classification: Typic Haplorthod

Parent material: Outwash sand

Physiography: Outwash plain

Slope: < 2 %

Comments: some logging about 40 years ago.

Oa

E

Bhs

B51

852

BC

0 - 5 cm. Black (N 2/0) well-decomposed hardwood litter; many fine roots;
abrupt wavy boundary.

5 - 15 cm. Grayish brown (10 YR 5/2) sand; weak medium subangular blocky
structure; very friable; many fine and medium roots; abrupt wavy

boundary.

15 - 20 cm. Dark reddish brown (5 YR 2.5/2) sand; weak medium subangular
blocky structure; very friable; common frne roots; dark reddish brown (5
YR 2.5/2) moderately cemented ortstein occurs in coltunns 10-15 cm
wide which extend into lower horizons and occupies about 40 percent of
the exposed surface; abrupt irregular boundary.

20 - 45 cm. Strong brown (7.5 YR 4/6) sand; weak medium subangular blocky
structure; friable; few fine roots; dark reddish brown (5 YR 2.5/2) very
strongly cemented ortstein occurs in columns 10 - 15 cm wide which

extend into lower horizons and occupies about 40 percent of the exposed
surface; clear wavy boundary.

45 - 75 cm. Strong brown (7.5 YR 5/6) sand; weak medium subangular blocky
structure; very fiiable; few fine roots; dark reddish brown (5 YR 2.5/2)
very strongly cemented ortstein occures in columns 10 - 15 cm wide
which exted into lower horizons and occupies about 40 percent of the
exposed surface; gradual wavy boundary.

75 - 105 cm. Yellowish brown (10 YR 5/4) sand; massive; strong brown (7.5
YR 4/6) moderately cemented ortstein occurs in columns 5 - 10 cm wide
and occupies about 10 percent of the exposed surface; gradual wavy
boundary.

189

C 105 - 150 cm. Light yellowish brown (10 YR 6/4) sand; massive.
C 210 - 230 cm.

Pedon: F OR-S

Location: SE 1/4 of SW 1/4 of SE 1/4 of SW 1/4, section 4, T 48 N, R 15 W
Date of description: August 16, 1994

Series: Rubicon

Subgroup classification: Entic Haplorthod

Parent material: Outwash sand

Physiography: Outwash plain

Slope: < 2%

Comments: Surface soil very dry.

Oi 0 - 4 cm. Dark brown (10 YR 3/3) partially decompoed hardwood and conifer
litter; many fine roots; abrupt wavy boundary.

E 4 - 19 cm. Grayish brown (10 YR 5/2) sand; single grain; loose; many fine and
common medium roots; abrupt wavy boundary.

B51 19 - 39 cm. Dark brown (7 .5 YR 4/4) sand; weak medium subangular blocky
structure; very friable; common fine roots; dark reddish brown (5 YR
3/3) moderately to strongly cemented ortstein occurs in columns 10 - 20
cm wide which extend into lower horizons and occupies about 30 percent
of the exposed surface; clear irregular boundary.

B52 39 - 69 cm. Strong brown (7.5 YR 4/6) sand; weak medium subangular blocky
structure; very friable; dark reddish brown (5 YR 3/4) moderately to
strongly cemented ortstein occurs in columns 10 - 20 cm wide which
extend into lower horizons and occupies about 30 percent of the exposed
surface; clear wavy boundary.

BC 69 - 89 cm. Yellowish brown (10 YR 5/4) sand; massive; strong brown (7.5 YR
4/6) moderately cemented ortstein occurs in columns 10 - 15 cm wide
and occupies about 10 percent of the exposed surface; clear wavy

boundary.
C 89 - 104 cm. Yellowish brown (10 YR 5/4) sand; massive.

C 240 - 250 cm.

Pedon: FOR-6

Location: NW 1/4 OfNE 1/4 OfNE 1/4 of SW 1/4, section 16, T 48 N, R 15 W

Date of description: August 12, 1994

Series: Deer Park

Subgroup classification: Spodic Udipsamment

Parent material: Outwash sand

Physiography: Outwash plain

Slope: < 2 %

Comments: Logging of large red and white pine about 10 - 20 years ago; charcoal
chunks in B horizon.

 

190

Oi 0 - 3 cm. Dark brown (10 YR 3/3) decomposing pine litter; abrupt wavy
boundary.

A 3 - 6 cm. Black (N 2/0) loamy sand; moderate medium granular structure; very
friable; many fine roots; abrupt wavy boundary.

E 6 - 15 cm. Grayish brown (10 YR 5/2) sand; weak medium subangular blocky

structure; very friable; many fine roots; abrupt wavy boundary.

B51 15 - 33 cm. Strong brown (7.5 YR 4/6) sand; weak medium subangular blocky
structure; very fiiable; many fine and common medium roots; dark brown
(7.5 YR 3/4) moderately cemented ortstein occurs in columns 10 - 20 cm
wide which extend into lower horizons and occupies about 10 percent of
the exposed surface; clear wavy boundary.

B52 33 - 58 cm. Yellowish brown (10 YR 5/6) slightly gravelly sand; weak coarse
subangular blocky structure; very fiiable; few fine roots; strong brown
(7.5 YR 4/6) weakly to moderately cemented ortstein occurs in columns
10 -20 cm wide and occupies about 10 percent of the exposed surface;

clear wavy boundary.

BC 58 - 78 cm. Yellowish brown (10 YR 5/4) sand; massive; clear wavy boundary.

C 78 - 98 cm. Light yellowish brown (10 YR 6/4) slightly gravelly sand; massive;
abrupt wavy boundary.

2C 98 - 108 cm. Brown (7.5 YR 5/4) loamy fine sand; massive; common medium
roots; abrupt wavy boundary.

3C 108 - 130 cm. Light yellowish brown (10 YR 6/4) sand; massive.

Pedon: FOR-7

Location: NE 1/4 of SE 1/4 of SE 1/4 of SW 1/4, section 18, T 48 N, R 15 W
Date of description: August 11, 1994

Series: Wallace

Subgroup classification: Typic Haplorthod

Parent material: Outwash sand

Physiography: Side of kettle depression in outwash; a ledge on the shoulder slope.
Slope: 5%

Comments: Some pit/mound topography in vicinity.

0a 0 - 4 cm. Black (10 YR 2/1) well decomposed hardwood and evergreen litter;
many fine roots; abrupt wavy boundary.

E 4 - 24 cm. Light brownish gray (10 YR 6/2) sand; weak medium subangular
blocky structure; very fiiable; many fine and common medium roots;
abrupt irregular boundary.

Bhs 24 - 30 cm. Dark reddish brown (5 YR 2.5/2) sand; weak medium subangular
blocky structure; friable; many fine roots; dark reddish brown (5 YR
2.5/2) moderately to very strongly cemented ortstein occurs in columns
10 - 20 cm wide which extend into lower horizons and occupies about 50
percent of the exposed surface; abrupt irregular boundary.

B51 30 - 49 cm. Strong brown (7.5 YR 4/6) sand; moderate medium subangular
blocky structure; fiiable; common fine roots; dark reddish brown and

5.5-.—

191

strong brown (5 YR 3/2 and 7.5 YR 4/6) strongly to very strongly
cemented ortstein occurs in columns 10 - 20 cm wide which extend into
lower horizons and occupies about 60 percent of the exposed surface;
clear irregular boundary.

B52 49 - 79 cm. Yellowish brown (10 YR 5/6) sand; weak medium subangular
blocky structure; very friable; few fine roots; dark reddish brown (5 YR
3/3) moderately to strongly cemented ortstein occurs in columns 10 - 20
cm wide and occupies about 40 percent of the exposed surface; clear
wavy boundary.

BC 79 - 119 cm. Yellowish brown (10 YR 5/4) sand; weak coarse subangular
blocky structure; very friable; clear wavy boundary.

C 124 - 140 cm. Light yellowish brown (10 YR 6/4) sand; massive.

C 119 - 124 cm.

C 220 - 240 cm.

Pedon: F OR-8

Location:

Date of description: August 10, 1994

Series: Wallace

Subgroup classification: Typic Haplorthod

Parent material: Outwash sand

Physiography: Outwash plain

Slope: < 2 %

Comments: Hemlock logging occurred about 40 years ago.

Oa

E

Bhsm

Bsml

Bsm2

2BC

3C

0 - 3 cm. Black (N 2/0) well decomposed hardwood litter; abrupt wavy
boundary.

3 - 14 cm. Grayish brown (10 YR 5/2) sand; weak medium subangular blocky
structure; very friable; many fine and medium roots; abrupt wavy

boundary.

14 - 19 cm. Dark brown (7.5 YR 3/2.5) sand; massive; few fine roots; very
strongly cemented ortstein occupies about 100 percent of the exposed
surface; abrupt irregular boundary.

19 - 43 cm. Strong brown (7.5 YR 4/6) sand; massive; few fine roots; very dark
brown (5 YR 2.5/2) moderately to very strongly cemented ortstein occurs
in columns 10 - 20 cm wide which extend to lower horizons and occupies
about 80 percent of the exposed surface; clear wavy boundary.

43 - 65 cm. Strong brown (10 YR 4/6) sand; massive; dark brown (7.5 YR 3/4)
moderately to strongly cemented ortstein occurs in columns 10 - 20 cm
wide and occupies about 50 percent of the exposed surface; abrupt wavy

boundary.

65 - 83 cm. Dark yellowish brown (10 YR 4/4) slightly gravelly coarse sand;
single grain; loose; clear wavy boundary.

83 - 103 cm. Yellowish brown (10 YR 5/4) sand; single grain; loose.

192

Pedon: FOR-9

Location:

Date of description: September 27, 1994
Series: Kalkaska

Subgroup classification: Typic Haplorthod
Parent material: Outwash sand
Physiography: Outwash plain

Slope: < 2%

Oi

E

Bhs

B51

B52

BC

C
C
C

0 - 5 cm. Black (N 2/0) partially decomposed hardwood litter; many very fine
roots; abrupt wavy boundary.

5 - 15 cm. Grayish brown (10 YR 5/2) sand; weak medium subangular blocky
structure; very friable; common fine roots; abrupt irregular boundary.

15 - 20 cm. Dark reddish brown (5 YR 2.5/2) sand; weak fine subangular blocky
structure; friable; many fine and medium roots; abrupt irregular
boundary.

20 - 35 cm. Dark brown (7.5 YR 4/4) sand; weak medium subangular blocky
structure; very friable; common medium roots; dark reddish brown (5 YR
3/2) strongly cemented ortstein occures in columns 15 - 20 cm wide
which extend into the horizons below and occupies about 40 percent of
the exposed surface; clear irregular boundary.

35 - 65 cm. Strong brown (7.5 YR 4/6) sand; weak medium subangular blocky
structure; friable; dark brown (7.5 YR 3/6) moderately to strongly
cemented ortstein occurs in columns 15 - 20 cm wide which extend into
the horizon below and occupies about 30 percent of the exposed surface;
clear wavy boundary.

65 - 90 cm. Yellowish brown (10 YR 5/4) sand; weak medium subangular
blocky structure; firable; dark brown (7.5 YR 4/4) moderately cemented
ortstein occurs in columns 15 - 20 cm wide and occupies about 10
percent of the exposed surface; clear wavy boundary.

90 - 120 cm. Yellowish brown (10 YR 5/4) sand; single grain; massive.
265 - 275 cm.
275 - 285 cm.

Pedon: SAV-l

Location: NE 1/4 of SE 1/4 OfNE 1/4, section 19, T 48 N, R 15 W

Date of description: July 14, 1994

Series: Rubicon

Subgroup classification: Entic Haplorthod

Parent material: Outwash sand

Physiography: Outwash plain

Slope: < 2 %

Comments: Scattered pit/mound pairs in this savanna. Stumps are plentiful.

B51

B52

BC

C

193

0 - 2 cm. Very dark gray (10 YR 3/ 1) sand. Weak very fine granular structure;
very friable; common fine roots; abrupt wavy boundary.

2 - 10 cm. Grayish brown (10 YR 5/2) sand. Single grain; loose; many fine
roots; clear wavy boundary.

10 - 45 cm. Dark brown (7.5 YR 4/4) gravelly sand. Weak fine subangular
blocky structure; very friable; common fine roots; dark reddish brown (5
YR 3/3) strongly cemented ortstein occurs in columns about 7 cm
extending into lower horizons and occupies about 20 percent of the
exposed surface; gradual wavy boundary.

45 - 80 cm. Dark yellowish brown (10 YR 4/4) slightly gravelly sand; single
grain; loose; few fine roots; strong brown (7.5 YR 5/6) weakly cemented
ortstein occurs in columns about 5 cm wide and occupies about 5 percent
of the exposed surface; gradual wavy boundary.

80 - 100 cm. Yellowish brown (10 YR 5/4) sand; single grain; loose; gradual
wavy boundary.

100 - 180 cm. Light yellowish brown (10 YR 6/4) sand; single grain; loose.

Pedon: SAV-2

Location: SW 1/4 of SE 1/4 of NE 1/4, section 20, T 48 N, R 15 W
Date of description: July 16, 1994

Series: Deer Park

Subgroup classification: Spodic Udipsamment

Parent material: Outwash sand

Physiography: Near edge of kettle in outwash plain, on shoulder slope.
Slope: < 2 %

Comments: Center of slight depression is about 100 m to the ENE.

A

B

B51

852

BC

0 - 2 cm. Very dark grayish brown (10 YR 3/2) sand; single grain; loose;
common fine roots; clear wavy boundary.

2 - 10 cm. Grayish brown (10 YR 5/2) sand; single grain; loose; common fine
roots; abrupt wavy boundary.

10 - 50 cm. Dark yellowish brown (7.5 YR 4/6) sand; weak medium subangular
blocky structure; very friable; few fine roots; dark reddish brown (5 YR
3/2) strongly cemented ortstein occurs in columns 5 - 8 cm wide which
extend into the horizon below and occupies about 20 percent of the
exposed surface; gradual wavy boundary.

50 - 85 cm. Dark yellowish brown (10 YR 4/6) slightly gravelly sand; weak
medium subangular blocky structure; very friable; dark brown (7.5 YR
3/4) moderately cemented ortstein occurs in columns about 5 cm wide;
gradual wavy boundary.

85 - 100 cm. Yellowish brown (10 YR 5/4) sand; massive; loose; gradual wavy
boundary.

100 - 170 cm. Pale brown (10 YR 6/3) sand; massive; loose.

194

Pedon: SAV-3

Location: NE 1/4 of SE 1/4 of NW 1/4, section 27, T 48 N, R 15 W

Date of description: July 20, 1994

Series: Deer Park

Subgroup classification: Spodic Udipsamment

Parent material: Outwash sand

Physiography: Outwash plain

Slope: < 2 %

A 0 - 3 cm. Very dark brown (10 YR 2/2) sand; weak fine granular structure; very
friable; many fine roots; clear wavy boundary.

E 3 - 13 cm. Grayish brown (10 YR 5/2) sand; weak fine subangular blocky
structure; very fiiable; many fine roots; clear wavy boundary.

B51 13 - 35 cm. Strong brown (7.5 YR 4/6) sand; weak fine subangular blocky
structure; very fiiable; common fine roots; dark reddish brown (5 YR
3/3) moderately to strongly cemented ortstein occurs in columns 5 - 8 cm
wide extending into lower horizons and occupies about 20 percent of the
exposed surface; gradual irregular boundary.

B52 35 - 65 cm. Yellowish brown (10 YR 5/4) sand; weak moderate subangular
blocky structure; very friable; few fine roots; strong brown (7.5 YR 5/6)
weakly to moderately cemented ortstein occurs in columns 5 - 8 cm wide
extending into lower horizons and occupies about 15 percent of the
exposed surface; gradual wavy boundary.

BC 65 - 90 cm. Light yellowish brown (10 YR 6/4) sand; massive; loose; strong
brown (7.5 YR 4/6) weakly to moderately cemented ortstein occurs in
columns 5 cm wide and occupies about 5 percent of the exposed surface;
gradual wavy boundary.

C 90 - 110 cm. Light yellowish brown (10 YR 6/4) sand; massive; loose.

Pedon: SAV-4

Location: SE 1/4 of NE 1/4 of SE 1/4, section 10, T 48 N, R 15 W
Date of description: July 26, 1994

Series: Rubicon

Subgroup classification: Entic Haplorthod

Parent material: Outwash sand

Physiography: Outwash plain

Slope: < 2 %

Comments: Recently burned stumps in area.

A 0 - 3 cm. Very dark gray (10 YR 3/1) sand; weak fine granular structure; very
fiiable; many fine roots; abrupt wavy boundary.

E 3 - 11 cm. Grayish brown (10 YR 5/2) sand; weak medium subangular blocky
structure; very friable; common fine roots; abrupt wavy boundary.

B51 11 - 30 cm. Dark brown (7.5 YR 4/4) slightly gravelly sand; weak medium
subangular blocky structure; very friable; common fine roots; dark

195

reddish brown (5 YR 3/2) moderately to strongly cemented ortstein
occurs in columns 5 - 8 cm wide extending into the horizons below and
occupying about 15 percent of the exposed surface; gradual irregular
boundary.

B52 30 - 60 cm. Strong brown (7.5 YR 5/6) sand; weak medium subangular blocky
structure; very friable; few fine roots; dark reddish brown (5 YR 3/4)
moderately cemented ortstein occurs in columns 5 - 8 cm wide extending
into the horizon below and occupies about 10 percent of the exposed
surface; gradual irregular boundary.

BC 60 - 90 cm. Yellowish brown (10 YR 5/4) sand; massive; strong brown (7.5 YR
4/6) weakly to moderately cemented ortstein occurs in columns 5 cm
wide and occupies abouto 5 percent of the exposed surface; gradual wavy
boundary.

C 90 - 120 cm. Yellowish brown (10 YR 5/4) sand; massive.

C 120 - 150 cm.

Pedon: SAV-S

Location: NW 1/4 of SE 1/4 of NW 1/4 of NE 1/4, section 17, T 48 N, R 15 W
Date of description: July 29, 1994

Series: Rubicon

Subgroup classification: Entic Haplorthod

Parent material: Outwash sand

Physiography: Kettle depression; backslope

Slope: 12 %

A

E

B51

B52

BC

0

0 - 3 cm. Very dark gray (10 YR 3/1) sand; weak fine granular structure; very
fiiable; many fine and medium roots; abrupt wavy boundary.

3 - 15 cm. Grayish brown (10 YR 5/2) sand; weak fine subangular blocky
structure; very friable; common fine and medium roots; abrupt wavy

boundary.

15 - 30 cm. Dark brown (7.5 YR 3/4) sand; weak fine subangular blocky
structure; very fiiable; few fine roots; dark reddish brown (5 YR 2.5/2)
weakly to moderately cemented ortstein occurs in columns 5 - 10 cm
wide and extends into lower horizons and occupies about 15 percent of
the exposed surface; clear wavy boundary.

30 - 55 cm. Dark yellowish brown (10 YR 4/6) sand; single grain; loose; few
fine roots; dark reddish brown (5 YR 3/3) weakly cemented ortstein
occurs in columns 5 - 8 cm wide and occupies about 5 percent of the
exposed surface; gradual wavy boundary.

55 - 75 cm. Yellowish brown (10 YR 5/4) sand; single grain; loose; gradual
wavy boundary.

75 - 120 cm. Light yellowish brown (10 YR 6/4) sand; massive.

240 - 260 cm.

196

Pedon: SAV-6

Location: NE 1/4 ofNW 1/4 OfNE 1/4 ofNW 1/4, section 9, T 48 N, R 15 W
Date of description: August 15, 1994

Series: Kalkaska

Subgroup classification: Typic Haplorthod

Parent material: Outwash sand

Physiography: Outwash plain

Slope: < 2 %

A 0 - 2 cm. Very dark gray (10 YR 3/1) sand; weak fine grandual structure; very
friable; many fine roots; abrupt wavy boundary.
E 2 - 12 cm. Grayish brown (10 YR 5/2) sand; weak fine subangular blocky

structure; very friable; many fine roots; abrupt irregular boundary.

Bhs 12 - 17 cm. Dark brown (7.5 YR 3/2) sand; weak fine subangular blocky
structure; very friable; many fine roots; abrupt irregular boundary.

B51 17 - 35 cm. Strong brown (7.5 YR 4/6) sand; weak medium subangular blocky
structure; very friable; common fine rOots; dark reddish brown (5 YR
3/2) moderately to very strongly cemented ortstein occurs in columns 15
- 20 cm wide extending into the horizon below and occupies about 30
percent of the exposed surface; clear wavy boundary.

B52 35 - 65 cm. Dark yellowish brown (10 YR 4/6) sand; weak medium subangular
blocky structure; very friable; few fine roots; dark reddish brown (5 YR
3/4) very strongly cemented ortstein occurs in columns 20 - 25 cm wide
extending into the horizon below and occupies about 40 percent of the
exposed surface; clear wavy boundary.

BC 65 - 95 cm. Yellowish brown (10 YR 5/4) sand; massive; dark reddish brown (5
YR 3/4) moderately to strongly cemented ortstein occurs in columns 15 -
20 cm wide and occupies about 20 percent of the exposed surface; clear

wavy boundary.
C 95 - 110 cm. Yellowish brown (10 YR 5/4) sand; massive.
C 220 - 240 cm.
Pedon: SAV-7

Location: NE 1/4 of NW 1/4 of SE 1/4 of SW 1/4, section 18,T 48 N, R 15 W
Date of description: August 9, 1994

Series: Rubicon

Subgroup classification: Entic Haplorthod

Parent material: Outwash sand.

Physiography: Outwash plain

Slope: < 2 %

A 0 - 6 cm. Very dark gray (10 YR 3/ 1) sand; weak medium granular structure;
fiiable; common fine roots; abrupt wavy boundary.
E 6 - 17 cm. Grayish brown (10 YR 5/2) sand; weak fine subangular blocky

structure; very friable; common fine roots; abrupt wavy boundary.

B51

B52

BC

C
C
C

197

17 - 35 cm. Reddish brown (7.5 YR 4/4) sand; weak fine subangular blocky
structure; very friable; common fine roots; dark reddish brown (5 YR
2.5/2) strongly cemented ortstein occurs in columns 8 - 12 cm wide
extending into the horizon below and occupies about 30 percent of the
exposed surface; clear irregular boundary.

35 - 70 cm. Yellowish brown (10 YR 5/6) sand; weak fine subangular blocky
structure; very friable; dark brown (7.5 YR 3/4) moderately to strongly
cemented ortstein occurs in columns 5 - 10 cm wide extending into the
horizon below and occupies about 25 percent of the exposed surface;
gradual wavy boundary.

70 - 110 cm. Yellowish brown (10 YR 5/4) sand; structureless; loose; strog
brown (7.5 YR 4/6) weakly cemented ortstein occurs in columns 5 - 10
cm wide and occupies about 10 percent of the exposed surface; gradual
wavy boundary.

110 - 150 cm. Yellowish brown (10 YR 5/4) sand; structureless; loose.
190 - 210 cm.
290 - 310 cm.

Pedon: SAV-8

Location: NE 1/4 of SE 1/4 of NW 1/4, section 10, T 48 N, R 15 W
Date of description: September 26, 1994

Series: Rubicon

Subgroup classification: Entic Haplorthod

Parent material: Outwash sand

Physiography: Outwash plain

Slope: < 2 %

A

85]

B52

0 - 8 cm. Very dark gray (10 YR 3/1) sand; weak fine subangular blocky
structure; very fiiable; many fine and very fine roots; abrupt wavy
boundary.

8 - 23 cm. Grayish brown (10 YR 5/2) sand; weak fine subangular blocky
structure; very fiiable; many fine roots; abrupt irregular boundary.

23 - 40 cm. Dark brown (7.5 YR 4/4) sand; weak medium subangular blocky
structure; very friable; common fine roots; dark reddish brown (5 YR
3/2) moderately to strongly cemented ortstein occurs in columns 10 - 15

cm wide extending into the horizon below and occupies about 30 percent
of the exposed surface; clear irregular boundary.

40 - 60 cm. Strong brown (7.5 YR 5/6) sand; weak medium subangular blocky
structure; friable; few fine roots; dark brown (7.5 YR 4/4) moderately to
strongly cemented ortstein occurs in columns 10 - 15 cm wide extending
into the horizon below and occupies about 20 percent of the exposed
surface; clear wavy boundary.

BC

198

60 - 90 cm. Yellowish brown (10 YR 5/4) sand; single grain; massive; dark
brown (7.5 YR 4/4) moderately cemented ortstein occurs in coltunns 15
cm wide and occupies about 10 percent of the exposed surface; clear
wavy boundary.

90 - 120 cm. Yellowish brown (10 YR 5/4) sand; single grain; massive.
275 - 282 cm.

LIST OF REFERENCES

Alban, D. H. 1982. Effects of nutrient accumulation by aspen, spruce, and pine on soil properties. Soil
Sci. Soc. Am. J. 46:853-861.

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

AuClair, A. N. 1976. Ecological factors in the development of intensive-management ecosystems in the
midwestem United States. Ecology 57:431-444.

Barnes, B. V., and W. H. Wagner, Jr. 1981. Michigan Trees: A Guide to the Trees of Michigan and the
Great Lakes Region. Ann Arbor: The University of Michigan Press. 383 pp.

Barrett, L. R., and R. J. Schaetzl. 1992. An examination of podzolization near Lake Michigan using
chronofunctions. Can. J. Soil Sci. 72:527-541.

Barrett, L. R., and R. J. Schaetzl. 1993. Soil development and spatial variability on geomorphic surfaces
of different age. Phys. Geog. 14:39-55.

Barrett, L. R., J. Liebens, D. G. Brown, R. J. Schaetzl, P. Zuwerink, T. W. Cate, and D.S. Nolan. 1995.
Relationships between soils and presettlement forests in Baraga County, Michigan. Am. Midl. Nat:
in press.

Barrow, C. J. 1991. Land Degradation: Development and Breakdown Of Terrestrial Environments.
Cambridge:Cambridge Univ. Press. 295 pp.

Beal, W. J. 1888. Forest Fires. In First report of the directors of the State Forestry Commission of
Michigan for the years 1887 and 1888. pp. 23-25.

Bell, R. E. 1975. A life of Russell Alexander Alger, 1836 - 1907. Unpublished Ph.D. Dissertation,
University of Michigan, Ann Arbor.

Benson, B. E. 1989. Logs and lumber: The development of the lumber industry in Michigan’s Lower
Peninsula, 1837-1870. Mount Pleasant:Central Michigan Univ. 309 pp.

Berg, R. C. 1984. The origin and early genesis of clay bands in youthful sandy soils along Lake
Michigan, USA. Geoderrna 32:45-62.

Bergquist, S. G. 1936. The Pleistocene history of the Tahquamenon and Manistique drainage region of
the northern peninsula of Michigan. Michigan Department of Conservation, Geological Survey
division. Publication NO. 40. pp. l-l48.

Blewett, W. L. 1994. Late Wisconsin history of Pictured Rocks National Lakeshore and vicinity. Pictured
Rocks Resource Report, Pictured Rocks National Lakeshore, Munising, MI. 8 pp.

Blewett, W. L., and R. L. Rieck. 1987. Reinterpretation of a portion of the Munising Moraine in northern
Michigan. Geol. Soc. Am. Bull. 98:169-175.

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

Bonham, C. D. 1989. Measurements for terrestrial vegetation. New York2Wiley and Sons. 338 pp.

Bourdo, E. A. 1956. A review of the General Land Office survey and of its use in quantitative studies of
former forests. Ecology 37:754-768.

Bourdo, E. A. 1983. The forest the settlers saw. In S.L. F lader (ed.), The Great Lakes forest: An
Environmental and Social History. Minneapolis: Univ. of Minnesota Press, pp. 17-32.

199

200

Brown, R. T., and J. T. Curtis. 1952. The upland conifer-hardwood forests of northern Wisconsin. Ecol.
Monog. 22:217-234.

Brown, R. T., and P. Mikola. 1974. The influence of fruticose soil lichens upon the mycorrhizae and
seedling growth of forest trees. Acta For. Fenn. 141:1-22.

Brubaker, L. B. 197 5. Postglacial forest patterns associated with till and outwash in northcentral Upper
Michigan. Quat. Res. 5:499-527.

Bruhn, J ., R. Heyd, D. Richter, and P. Flemington. 1987. Lichen control and mycorrhizal studies in pine
plains reforestation efforts. In B. A. Montgomery (ed.), Michigan Forest Pest Report 1987. Mich.
State Univ., E. Lansing. pp. 60-64.

Bryan, K., and C. C. Albritton, Jr. 1943. Soil phenomena as evidence of climatic changes. Am. J. Sci.
241:469-489.

Brydon, J. E. 1965. Clay illuviation in some Orthic Podzols of Eastern Canada. Can. J. Soil Sci. 45: 127-
138.

Buol, S. W., F. D. Hole, and R. J. McCracken. 1980. Soil Genesis and Classification. Ameszlowa State
Univ. Press. 406 pp.

Burton, C. M., and M. A. Burton. 1930. History of Wayne County and the City of Detroit, Michigan.
Vol. 11. Chicago: Clarke Publishing Co.

Buttrick, P. L. 1921. A study of regeneration on certain cut-over hardwood lands in northern Michigan. J.
Forestry 19:872-876.

Carey, L. M. 1993. Soil Survey of Pictured Rocks National Lakeshore. United States Department of
Agriculture, Soil Conservation Service, US. Govt. Printing Office, Washington, DC.

Carter J. L. 1967. Voyageurs’ Harbor. Pilot Press: Grand Marais. 66 pp.

Carter, J. R. 1988. Digital representations of topographic surfaces. Photog. Eng. and Rem. Sens. 54: 1577-
1580.

Canham, C. D., and O. L. Loucks. 1984. Catastrophic windthrow in the presettlement forests of
Wisconsin. Ecology 65:803-809.

Cayford, J. H., and D. J. McRae. 1983. The ecological role of fire in jack pine forests. In R. W. Wein and
D. A. MacLean (eds.) The Role of Fire in Northern Circumpolar Ecosystems, pp. 183-199. New
York:Wiley & Sons.

Clark, J. S. 1989. Ecological disturbance as a renewal process: Theory and application to fire history.
Oikos 56:17-30.

Cottam, G., and J. T. Curtis. 1956. The use of distance measures in phytosociological sampling. Ecology
37:451-460.

Curtis, J. T. 1971. The Vegetation of Wisconsin: An Ordination of Plant Communities. Madisoanniv.
of Wisconsin Press. 657 pp.

Dahlgren, R. A., and F. C. Ugolini. 1991. Distribution and characterization of short-range-order minerals
in Spodosols from the Washington Cascades. Geodenna 48:391-413.

Daly, B. K. 1982. Identification of Podzol and Podzolized soils in New Zealand by relative absorbence of
oxalate extracts of A and B horizons. Geodenna 28:29-38.

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

DeKimpe, C. R., and Y. A. Martel. 1976. Effects of vegetation on the distribution of carbon, iron, and
aluminum in the B horizons of northern Appalachian Spodosols. Soil Sci. Soc. Am. J. 40:77-80.

Delcourt, H. R., and P. A. Delcourt. 1974. Primeval magnolia-holly-beech climax in Louisiana. Ecology
55:638-644.

201

Delcourt, H. R., and P. A. Delcourt. 1977. Presettlement magnolia-beech climax of the gulf coastal plain:
Quantitative evidence from the Apalachicola River Bluffs, North-Central Florida. Ecology 58: 1085-
1093.

Drexler, C. W., W. R. F arrand, and J. D. Hughes. 1983. Correlation of glacial lakes in the Superior Basin
with eastward discharge events from Lake Agassiz. In J. T. Teller and L. Clayton (eds.) Glacial Lake
Agassiz. Geol. Assoc. of Canada Special Paper 26. pp. 309-329.

Elliot, J. C. 1953. Composition of upland second growth hardwood stands in the tension zone of Michigan
as affected by soils and man. Ecol. Monog. 231271-287 .

Evans, L. J ., and B. H. Cameron. 1985. Color as a criterion for the recognition of podzolic B horizons.
Can J. Soil Sci. 65:363-370.

Evans, L. J., and W. G. Wilson. 1985. Extractable Fe, A1, Si, and C in B horizons of podzolic and
brunisolic soils from Ontario. Can. J. Soil Sci. 65:489-496.

Farmer, V. C. 1984. Distribution of allophane and organic matter in podzol B horizons: Reply to
Buurman and van Reeuwijk. J. Soil Sci. 35:453-458.

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

Fisher, R. F. 1979. Possible allelopathic effects of reindeer-moss (Cladonia) on jack pine and white
spruce. Forest Sci. 25:256-260.

Frelich, L. E., and C. G. Lorimer. 1991. Natural disturbance regimes in hemlock-hardwood forests of the
upper Great Lakes region. Ecol. Monog. 61 : 145-164.

Fralish, J. S., F. B. Crooks, J. L. Chambers, and F. M. Harty. 1991. Comparison of presettlement, second-
growth and old-grth forest on six site types in the Illinois Shawnee Hills. Am. Midl. Nat. 1251294-
309.

Franzmeier, D. P. 1962. A chronosequence of Podzols in Northern Michigan. Unpublished Ph.D.
dissertation, Michigan State University, East Lansing. Michigan.

Franzmeier, D. P., and E. P. Whiteside. 1963. A chronosequence of Podzols in Northern Michigan. 11.
Physical and chemical properties. Mich. Quart. Bull. 46:20-36.

Frederick, D. J ., L. Rakestraw, C. R. Edier, and R. A. Van Dyke. 1976. Original forest vegetation of the
Pictured Rocks National Lakeshore and a comparison with present conditions. Mich. Acad. 9:43 3-
443.

Fries, R. F. 1951. Empire in Pine: The Story of Lumbering in Wisconsin 1830-1900. Madison: The State
Historical Society of Wisconsin. 285 pp.

Futyma, R. P. 1981. The northern limits of glacial Lake Algonquin in Upper Michigan. Quat. Res.
15:291-310.

Gleason, H. A. 1918. On the development of two plant associations of northern Michigan. Plant World
21:151-158.

Graham, S. A. 1941. Climax forests of the Upper Peninsula of Michigan. Ecology 22:355-362.
Grieg-Smith, P. 1983. Quantitative Plant Ecology. OxfordzBlackwell Scientific Publications. 359 pp.

Grier, C. C. 1975. Wildfire efi‘ects on nutrients distribution and leaching in a coniferous ecosystem. Can.
J. For. Res. 5:599-607.

Grimm, E. C. 1984. Fire and other factors controlling the big woods vegetation of Minnesota in the mid-
nineteenth century. Ecol. Monogr. 54:291-311.

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

Hannah, P. R., and R. Zahner. 1970. Nonpedogenetic texture bands in outwash sands of Michigan: Their
origin, and influence on tree growth. Soil Sci. Soc. Am. Proc. 34:134-136.

202

Heinselman, M. L. 1973. Fire in the virgin forests of the Boundary Waters Canoe Area, Minnesota. Quat.
Res. 32329-382.

Hole, F. D. 1975. Some relationships between forest vegetation and Podzol B horizons in soils of
Menominee tribal lands, Wisconsin, USA. Sov. Soil Sci. 7:714-23.

Holling, C. C. 1973. Resilience and stability of ecological systems. Ann. Rev. Ecol. Syst. 4: 1-23.

Host, G. E., and K. S. Pregitzer. 1992. Geomorphic influences on ground-flora and overstory composition
in upland forests of northwestern lower Michigan. Can. J. For. Res. 22: 1547-1555.

Host, G. E., K. S. Pregitzer, C. W. Ramm, J. B. Hart, and D. T. Cleland. 1987. Landforrn-mediated
differences in successional pathways among upland forest ecosystems in northwestern Lower
Michigan. For. Sci. 33:445-457.

Host, G. E., K. S. Pregitzer, C. W. Ramm, D. P. Lusch, and D. T. Cleland. 1988. Variation in overstory
biomass among glacial landforms and ecological land units in northwestern Lower Michigan. Can. J.
For. Res. 18:659-668.

Hotchkiss, G. W. 1898. History of the Lumber and Forest Industry of the Northwest. ChicagozG. W.
Hotchkiss & Co.

Huang, P. M. 1991. Ionic factors affecting the formation of short-range ordered aluminositicates. Soil
Sci. Soc. Am. J. 55:1172-1180.

Hushen, T. W., R. O. Kapp, R. D. Bogue, and J. T. Worthington. 1966. Presettlement forest patterns in
Montcalm County, Michigan. Mich. Bot. 51192-211.

Inoue, K., and P. M. Huang. 1986. Influence of selected organic ligands on the formation of allophane
and imogolite. Soil Sci. Soc. Am. J. 50:1623-1633.

Jauhiainen, E. 1973. Age and degree of podzolization of sand soils on the coastal plain of northwest
Finland. Comment. Biologicae 69: 1-32.

Jersak, J., R. Amundson, and G. Brimhall, Jr. 1995. A mass balance analysis of podzolization: examples
from the northeastern United States. Geodenna 66:15-42.

Johnson, D. L., and D. Watson-Stegner. 1987. Evolution model of pedogenesis. Soil Sci. 143:349-366.

Kittredge, J., and A. K. Chittenden. 1929. Oak forests of northern Michigan. Michigan Agricultural
Experiment Station Special Bulletin 190.

Kodama, H., and C. Wang. 1989. Distribution and characterization of noncrystalline inorganic
components in Spodosols and Spodosol-like soils. Soil Sci. Soc. Am. J. 53:526-534.

Koutaniemi, L., R. Koponen, and K. Rajanen. 1988. Podzolization as studied from terraces of various
ages in two river valleys, Northern Finland. Silva F ennica 22:113-133.

Lee, F. Y., T. L. Yuan, and V. W. Carlisle. 1988a. Nature of cementing materials in ortstein horizons of
selected Florida Spodosols: I. Consituents of cementing materials. Soil Sci. Soc. Am. J. 52: 141 1-
1418.

Lee, F. Y., T. L. Yuan, and V. W. Carlisle. 1988b. Nature of cementing materials in ortstein horizons of
selected Florida Spodosols: 11. Soil properties and chemical form(s) of aluminum. Soil Sci. Soc. Am.
J. 52:1796-1801.

Leitner, L. A., C. P. Dunn, G. R. Guntenspergen, F. Steams, and D. M. Sharpe. 1991. Effects of site,
landscape features, and fire regime on vegetation patterns in presettlement southem Wisconsin.
Landscape Ecol. 5:203-217.

Likens, G. E., H. Bormann, R. S. Pierce, J. S. Eaton, and N. M. Johnson. 1977. Biogeochemistry of a
Forested Ecosystem. Springer-Verlag: New York. 146 pp.

Little, S. 1974. Effects of fire on temperate forests: Northeastern United States. In T. T. Kozlowski and
C. E. Ahlgren (eds.) Fire and Ecosystems. New Yorszcademic Press. pp. 225-250.

203

Livingston, B. E. 1905. The relation of soils to natural vegetation in Roscommon and Crawford Counties.
Bot. Gaz. 39:22-41.

Loope, W. L. 1991. Interrelationships of fire history, land use history, and landscape pattern within
Pictured Rocks National Lakeshore, Michigan. Can. Field-Naturalist 105:18-28.

Maissurow, D. K. 1941. The role of fire in the perpetuation of virgin forests in northern Wisconsin. I.
Forestry 39:201-207.

Marquis, D. A. 1990. Prunus serotina Ehrh. Black Cherry. In Burns, R. M., and B. H. Honkala (eds.)
Silvics of North America. Volume 2, Hardwoods. Agriculture Handbook 654. United States Forest
Service, United States Govt. Printing Office, Washington, D. C., USA. pp. 594-604.

McKeague, J. A. (ed.) 1978. Manual on Soil Sampling and Methods of Analysis. 2nd ed.
OttawazCanadian Society of Soil Science.

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

McKeague, J. A., and J. H. Day. 1966. Dithionite- and oxalate-extractable Fe and A] as aids in
differentiating various classes of soils. Can. J. Soil Sci. 46:13-22.

McKeague, J. A., and C. Wang. 1980. Micromorphology and energy dispersive analysis of ortstein
horizons of podzolic soils from New Brunswick and Nova Scotia, Canada. Can J. Soil Sci. 60:9-21.

Michigan Railroad Commission. 1919. Aids, Gifis, Grants and Donations to Railroads Including Outline
of Development and Successions in Titles to Railroads in Michigan. Wynkoop Hallenbeck Crawford
Co: Lansing, Michigan.

Miles, 1. 1985. The pedogenic effects of different species and vegetation types and the implications of
succession. J. Soil Sci. 36:571-584.

Miles, R. J., and D. P. Franzmeier. 1981. A lithochronosequence of soils formed in dune sand. Soil Sci.
Soc. Am. J. 45:362-367.

Mitchell, J. 1973. The bracken problem. In J. Tivy (ed.) The Organic Resources of Scotland: Their
Nature and Evaluation. Oliver and Boyd:Edinburgh. pp. 98 - 108

Mizota, C. 1982. Clay mineralogy of spodic horizons from Cryorthods in Japan. Soil Sci. Plant Nutr.
28:257-268.

Mokma, D. L. 1993. Color and amorphous materials in Spodosols from Michigan. Soil Sci. Soc. Am. J.
57:125-138.

Mokma, D. L., and G. F. Vance. 1989. Forest vegetation and origin of some spodic horizons, Michigan.
Geoderma 43:31 1-324.

Mokma, D. L., R. J. Schaetzl, J. A. Doolittle, and E. P. Johnson. 1990. Ground-penetrating radar study of
ortstein continuity in some Michigan Haplaquods. Soil Sci. Soc. Am. J. 54:936-938.

Muhs, D. R. 1984. Intrinsinc thresholds in soil systems. Phys. Geog. 5:99-110.

Nomberg, P. 1977. Soil profile development in sands of varying age in Vendsyssel, Denmark. Catena
4: 165-179.

Narnberg, P. 1980. Mineralogy of a podzol formed in sandy materials in northern Denmark. Geodenna
24:25-43.

Palik, B. J., and K. S. Pregitzer. 1992. A comparison of presettlement and present-day forests on two
bigtooth aspen-dominated landscapes in Northern Lower Michigan. Am. Midl. Nat. 127:327-338.

Parfitt, R. L., and C. W. 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.

Parfitt, R. 1.., and J. M. Kimble. 1989. Conditions for formation of allophane in soils. Soil Sci. Soc. Am.
J. 53:971-977.

204

Phillips, E. A. 1959. Methods of Vegetation Study. New York: Holt, Rinehart, and Wintson. 107 pp.

Phillips, J. D. 1993a. Progressive and regressive pedogenesis and complex soil evolution. Quat. Res.
40:169-176.

Phillips, .1. D. 1993b. Stability implications of the state factor model of soils as a nonlinear dynamical
system. Geoderma 5821-15.

Potzger, J. E., M. E. Potzger, and J. McCormick. 1956. The forest primeval of Indiana as recorded in the
original U.S. land surveys and an evaluation of previous interpretations of Indiana vegetation. Butler
Univ. Bot. Stud. 13:95-111.

Quinby, P. A. 1991. Self-replacement in old-growth white pine forests in Temagami, Ontario. Forest
Ecology and Management 41:95-109.

Quinn, M.-L. 1991. The Appalachian Mountains’ Copper Basin and the concept of environmental
susceptibility. Environ. Manage. 15:179-194.

Ranger, J ., E. Dambrine, M. Robert, D. Righi, and C. Felix. 1991. Study of current soil-fonning processes
using bags of vermiculite and resins placed within soil horizons. Geodenna 48:335-350.

Ranger, J ., and C. Nys. 1994. The effect of spruce (Picea abies Karst.) on soil development: An
analytical and experimental approach. Eur. J. Soil Sci. 45:193-204.

Rector, W. G. 1953. Log Transportation in the Lake States Lumber Industry, 1840 - 1918: The
Movement of Logs and its Relationship to Land settlement, Waterway Development, Railroad
Concentration, Lumber Production, and Prices. American Waterways Series no. 4. Glendale,
Calif.:Arther H. Clark.

Righi, D., S. Bravard, A. Chauvel, J. Ranger, and M. Robert. 1990. In situ study of soil processes in an
Oxisol-Spodosol sequence of Amazonia (Brazil). Soil Sci. 150:438-445.

Rohe, R. E. 1971. Lumbering’s impact on the landscape of the Wolf River area of northeastern
Wisconsin. Unpublished M. A. thesis, University of Colorado.

Roth, F. 1898. Forestry conditions of northern Wisconsin. Wisconsin Geological and Natural History
Survey Bulletin no. 1. Madison, Wisconsin. 78 pp.

Roth, F. 1906. Biennial report of forest warden. In Michigan Forestry Commission, Report of the
Michigan Forestry Commission for the years 1905-1906. pp. 15-49.

Roth, F. 1908. Biennial report of forest warden. In Michigan Forestry Commission, Report of the
Michigan Forestry Commission for the years 1907-1908. Lansing, Michigan. pp. 16-32.

Russell, E. W. B. 1983. Indian-set fires in the forests of the northeastern United States. Ecology 64:78-
88.

Safford, L. O., J. C. Bjorkbom, and J. C. Zasada. 1990. In Burns, R. M., and B. H. Honkala (eds.) Silvics
of North America. Volume 2, Hardwoods. Agriculture Handbook 654. United States Forest Service,
United States Govt. Printing Office, Washington, D. C., USA. pp. 158 - 171.

Sawyer, A. L. 1919. The forests of the Upper Peninsula and their place in history. Mich. Hist. 3:367-383.

Schaetzl, R. J. 1990. Effects of treethrow microtopography on the characteristics and genesis of
Spodosols, Michigan, USA. Catena 17:111-126.

Schaetzl, R. J. 1992. Texture, mineralogy, and lamellae development in sandy soils in Michigan. Soil Sci.
Soc. Am. J. 56:1538-1545.

Schaetzl, R. J ., and D. G. Brown. n.d. Forest species gradients and associations in Baraga County,
Michigan, ca. 1850. Great Lakes Geographer: under review.

Schaetzl, R. J., and S. A. Isard. 1990. Comparing “warm season” and “snowmelt” pedogenesis in
Spodosols. In J. M. Kimble and W. D. Nettleton, eds., Proceedings of the Fifth International Soil
Correlation Meeting (ISCOM 1V) Characterization, Classification, and Utilization of Spodosols.
USDA, Soil Conservation Service, Lincoln, NE. pp. 303-318.

205

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

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

Sheldrick, B. H. (ed.). 1984. Analytical Methods Manual. Land Resource Research Institute Contrib. No.
84-30. Ottawa, Canada.

Sherrard, T. H. 1903. The Michigan Forest Reserve. Report of the Michigan Forestry Commission for
the year 1902. pp. 28-54.

Shetron, S. G. 1972. Forest site productivity among soil taxonomic units in Northern Lower Michigan.
Soil Sci. Soc. Am. Proc. 36:358-363.

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.

Shoji, S., and H. Yamada. 1991. Comparisons of mineralogical properties between tephra-derived
Spodosols from Alaska and nontephra-derived Spodosols from New England. Soil Sci. 152: 162-183.

Sirnard, A. J. and R. W. Blank. 1982. Fire history of a Michigan jack pine forest. Mich. Acad. 15:59-71.

Singer, M. J., and P. Janitzky. 1986. Field and laboratory procedures used in a soil chronosequence study.
USGS Bulletin 1648. US Govt. Printing Office, Washington, DC. 49 pp.

Singer, M., F. C. Ugolini, and]. Zachara. 1978. In situ study of podzolization on tephra and bedrock.
Soil Sci. Soc. Am. J. 42:105-111.

Skeels, F. C. 1898. Present condition of forests and stump land. Mich. Agr. Coll. Exp. Sta. Bull. no. 162.
pp. 7-12.

Soil Survey Staff. 1981. Soil Survey Manual. US. Department of Agriculture-Soil Conservation Service.
US. Govt. Printing Office, Washington, DC.

Soil Survey Laboratory Stafi‘. 1992. Soil Survey Laboratory Methods Manual. Soil Survey Investigations
Report No. 42, Version 2.0. US. Department of Agriculture, Soil Conservation Service, National Soil
Survey Center, Lincoln, Nebraska.

Sparhawk, W. N., and W. D. Brush. 1929. The economic aspects of forest destruction in northern
Michigan. USDA Tech. Bull. no. 92. US. Govt. Printing Office, Washington, DC.

Stanley, S. R., and E. J. Ciolkosz. 1981. Classification and genesis of Spodosols in the Central
Appalachians. Soil Sci. Soc. Am. J. 45:912-917.

Swain, A. M. 1973. A history of fire and vegetation in northeastern Minnesota as recorded in lake
sediments. Quat. Res. 32383-396.

Takahashi, T., T. Fukuoka, and R. A. Dahlgren. 1995. Aluminum solubility and release rates from soil
horizons dominated by aluminum-humus complexes. Soil Sci. Plant Nutr. 41:119-131.

Twining, C. E. 1983. The lumbering frontier. In S. L. Flader (ed.) The Great Lakes forest: An
Environmental and Social History. MinneapoliszUniv. of Minnesota Press. pp. 121-136.

Ugolini, F. C., R. Dahlgren, J. LaManna, W. Nuhn, and J. Zachara. 1991. Mineralogy and weathering
processes in Recent and Holocene tephra deposits of the Pacific Northwest, USA. In M. J. Pavich, ed.,
Weathering and Soils. Geodenna 51:277-299.

Ugolini, F. C., R. Dahlgren, S. Shoji, and T. Ito. 1988. An example of andosolization and podzolization as
revealed by soil solution studies, southern Hakkoda, northeastern Japan. Soil Sci. 145:111-125.

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

Ugolini, F. C., and R. S. Sletten. 1991. The role of proton donors in pedogenesis as revealed by soil
solution studies. Soil Sci. 151:59-75.

Van Wagner, C. E. 1970. Fire and red pine. Proceedings Tall Timbers Fire Ecol. Conf. 10:211-219.

206

Veatch, J. O. 1928. Reconstruction of forest cover based on soil maps. Quart. Bull. Mich. Ag. Exp. Sta.
10:116-126.

Veatch, J. O., L. R. Schoenmann, F. R. Lesh, and Z. C. Foster. 1929. Soil Survey of Alger County,
Michigan. United States Department of Agriculture Bureau of Chemistry and Soils. Washington, DC.
41 pp.

Vogl, R. J. 1964a. The effects of fire on the vegetational composition of bracken-grasslands. Trans. Wisc.
Acad. Sci. Arts Let. 53:67-82.

Vogl, R. J. 1964b. Vegetational history of Crex Meadows, a prairie savanna in northwestern Wisconsin.
Am. Midl. Nat. 72:157-175.

Vogl, R. J. 1970. Fire and the northern Wisconsin pine barrens. Proceedings Tall Timbers Fire Ecol.
Conf. 10:175-209.

Walters, R. S., and H. W. Yawney. 1990. Acer rubrum L. Red Maple. In Burns, R. M., and B. H.
Honkala (eds.) Silvics of North America. Volume 2, Hardwoods. Agriculture Handbook 654.
United States Forest Service, United States Govt. Printing Office, Washington, D. C., USA. 877 pp.

Wang, C., G. J. Beke, and J. A. McKeague. 1978. Site characteristics, morphology and physical properties
of selected ortstein soils from the Maritime provinces. Can. J. Soil Sci. 58:405-420.

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

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

Wang, C., and H. W. Rees. 1980. Characteristics and classification of non-cemented sandy soils in New
Brunswick. Can. J. Soil Sci. 60:71-81.

White, C. A. 1984. A History of the Rectangular Survey System. United States Department of the
Interior, Bureau of Land Management, Washington, DC. 774 pp.

Whitney, G. G. 1986. Relation of Michigan's presettlement pine forests to substrate and disturbance
history. Ecology 67:1548-1559.

Whitney, G. G. 1987. An ecological history of the Great Lakes forest of Michigan. J. Ecol. 75:667-684.

Whitney, G. G. 1990. The history and status of the hemlock-hardwood forests of the Allegheny Plateau.
J. Ecol. 78:443-458.

Wilde, S. A. 1933. The relation of soils and forest vegetation of the lakes states region. Ecology 14:94-
105.

Williams, M. 1989. Americans and their Forests: A Historical Geography. Cambridge:Cambridge
University Press. 599 pp.

HICH 6

11111111111111“