VARIABILITY OF THE PRODUCT IVE CAPACITY
OF RUBICON SAND FOR RED PINE -

Thesis for the Degree of M. S.
MICHIGAN STATE UNIVERSITY
RONALD JULIUS CHURCH
1977

- w
L [BI {Y
dick"; . Late

Univcrxfiy

 
   
  

  

 

 

ABSTRACT

VARIABILITY OF THE PRODUCTIVE CAPACITY
OF RUBICON SAND FOR RED PINE

By
Ronald Julius Church

Soil classification at the series level has been used by
foresters as a guide in estimating the productive capacity of
the land. Since today's forests must be managed intensively to
obtain the maximum production from the land, foresters prefer to
be able to predict the productive capacity of a given site, as
measured by site index, to within a reasonably narrow range such
as one site class, or t 5 feet. The site index of red pine in
plantations growing on the Rubicon soil series in northern lower
Michigan has been reported to be 58 feet at 50 years with a range,ie.
one standard deviation, of i 7 feet. This means that the pro-
ductive capacity of the land is not adequately estimated in term's
of intensive forest management when the Rubicon soil series is
used as a guide. This thesis therefore, is an investigation of
the variability of the productive capacity of Rubicon sand for
red pine plantations.

The study area was confined to two counties in the north-
western portion of the lower peninsula of Michigan. The landforms
in the area are mainly outwash plains and moraines, which are
Wisconsinan in age.

Site index curves were developed from stem analysis data in
this study. Harmonized site index curves were based on the fol-

lowing equation that was developed: (site index) = -H.Ol7 +

1.399(age) + .005197(age)2. Polymorphic site index curves were
also generated but did not result in important differences in terms
of management objectives.

The site index for each plot was determined from the above
mentioned curves and from other published curves which represented
broader geographical ranges and/or native stands of red pine.

It was found that various site index curves yield similar site
index values for red pine on Rubicon sand. However, the varia-
tion in site index between plots ranged from 7.8% to 11.8%, with
the different curves,or an average of 10.2%.

In a similar fashion, volume at age 50 was used to estimate
the productive capacity of the land. Vblume was a_more sensitive
indicator than site index. Like site index however, the average
volume as determined by different methods was similar but the
variation between plots was greater and ranged from 11.3% to 19.4%.

Stepwise regression analysis was employed to find meaningful
correlations between site index for red pine and numerous proper-
ties of Rubicon sand and non-soil site factors. The site factors
which were associated with 71% of the variation in site index are:
pH of the B22 horizon, thickness of the A horizon, basal area,
percent gravel in the profile, and average percent fine sand of
the A.p1us B horizons. Thus, these factors are apparently the
most important characteristics of Rubicon sand known to be corre-
lated with variations in site index for red pine. Until more is
known about the site factors associated with more of the variation
in red pine site index, or variation in Rubicon sand, there is
little hope for improving on soil classification for red pine

site quality estimation on this soil.

VARIABILITY OF THE PRODUCTIVE CAPACITY
OF RUBICON SAND FOR RED PINE
By
Ronald Julius Church

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE

Department of Crop and Soil Sciences

1977

ACKNOWLEDGEMENTS

Throughout the course of this study I have called upon the
help and guidance of many people. At this time I would like to
thank everyone, especially Dr. E. P. Whiteside for his enthusiasm
and encouragement during this study. He has made me more aware
of the land around me and has given me the tools with which I can
better understand it.

Further, gratitude is due the Forestry Division of Michigan's
Department of Natural Resources for the use of their records and
equipment, plus their advice, encouragement and financial assist-
ance. I am also indebted to the United States Forest Service
personnel for their time and assistance.

I would also like to give Special thanks to:

Dan Sandahl for his invaluable assistance at each plot and

his moral support;

My guidance committee for their help throughout the course

of this study;

Mary, for her patience and understanding; and

Last, but not least, my Mother and Father, without whom

none of this would have been possible.

ii

.\. at

In QIA vl‘

‘. V d a n \ BIN
T s
W h
N3
Ex
R.

TABLE OF CONTENTS

Introduction .

Background . . . . . . . . . . . .
Soil Classification System . . . . . . .
Site . . . . . . . . . . . . . . . . .

Literature Review

Importance of Site Quality

Total Site Factors Affecting Site Quality .
Individual Factors Affecting Site Quality .

Study area and Procedures

Nature of the Study Area
Geology
Climate . . . . . . .
vegetation . . . . . . . .
Soils .

Selection of Sites

Field Procedures

Laboratory Procedures .

Computor Analyses .

Volume Determinations .

Results and Discussion . . . . . . . .
Tree Data . . . . . . . . . . . .
Soil Data .
Soil Texture . . . .

Depth to Water Table .

Gravel Content . . . . .
Landforms . . . . . . . . .
Soil Nutrients . . . .

iii

A Page

10
10
13
19
2a
2a
2a
26
27
27
28
29
3o
30
33
36
35
58
59
59
50

61
62

Conclusions . . . . . . . . . . . . . . .
List of References . . . . . . . . . . . . . .
Appendices . . . . . . . . . . . . . .

I Rubicon and Graycalm Soil Series

II

III

Descriptions . . . . .
Table A. Coefficients of Determination
for all Equations Tested . . . .
Table B. Summary of the Tests for
Linearity. .

Table C. Sumary of Analyses of Covariance.
Testing the Difference between Levels and
Slopes of the Individual Plot Equations
with that of the Overall Mean Equation . .
Table D. Summary of the Analyses of
Covariance of the Mem Curve for the
Faster Growing Plots and the Mean Curve
for the Slower Growing Plots when Compared
to their Respective Individual Plots .
Table E. Comparison of Individual Plot
Growth Curves to the Mean Growth Curve of
All Plots Combined . . .
Variables Used in Stepwise Regression
Analyses . . . . . . . .
Notes Taken at Each Plot. . . . . . .
Summary of Laboratory Analysis Data . . . .

iv

Page
70
73

78
82

8Q

85

86

87

97
98
122

 

II

 

III

II

LIST OF TABLES

Tables Page

I Summary of Equations Developed from the
Pooled Data of Tree Height and Age . . . . . . . 38

II Summary Table for the Equations of the Height
and Age of the Faster and Slower Growing Groups
of Plots . . . . . . . . . . . . . . . . . . . #1

III The "b" Coefficients, or Slope Constants, of
the Straight Line Equations for the Individual
Plots . . . . . . . . . . . . . . . . . . . #8

IV Comparison of Red Pine Site Index Values of
Plots obtained from Curves Developed for This ,
and Other Studies . . . . . . . . . . #9

V Comparison of Estimated Cubic Foot volume of
Red Pine at 50 Years of Age . . . . . . . . . 55

VI Summary of Available and Exchangeable Nutrients
of Rubicon Sand Horizons. These values are
the average of all 10 plots . . . . . . . . . . 6%

VII Summary of Stepwise Regression Analyses;
variables chosen at each step . . . . . . . . . 66

Figure

LIST OF FIGURES

Outline of the TWO County Study Area, its
Surface Geology and Plot Locations .

Scatter Diagram of Heights and Ages of All
Red Pine Trees Sampled on Rubicon Sand .

Mean Growth Curve for the Faster and Slower
Growing Plots . . . ... . . . . .

Harmonized Site Index Curves Using the Mean
Growth Curve of All 10 Plots Combined as
the Guiding Curve . . . . . . . . . . . .

Polymorphic Site Index Curves. The Mean
Growth Curve of the Faster Growing Plots
was Used as the Guiding Curve for Site
Index Classes 70 and 60. The Mean Growth
Curve of the Slower Growing Plots was Used
as the Guide Curve for Site Index Classes
50 and no . . . . . . . . . . . . . .

Comparison of Polymorphic and Harmonized
Curves . . . . . . . . . . . . . . . . .

Comparison of the Mean Growth Curves from
This and Other Studies . . . . . . .

Comparison of the Polymorphic Site Index
Curves Developed for This and van Eek's
Study . . . . . . . . . . . . . . . .

'Harmonized Vblume Growth Curves Using the

Mean Growth Curve of All Trees Combined

‘as the Guiding Curve . . .

vi

Page
25
37

42

an

”5
H6

51
53

56

INTRODUCTION

Red pine (Pings resinosa Ait.) is an economically important
species and has been planted extensively throughout the northern
half of the lower peninsula of Michigan. The Department of Natural
Resources of Michigan has planted about 100,000 acres‘. The site
index of red pine growing on Rubicon sand (Entic Haplorthod) has
been reported to be 58 3 7 feet (van Eck, 1958). As an estimate
of the productive capacity of Rubicon sand, this site index is
quite variable. If the site index were within one site class,
that is, if it varied less than t 5 feet, it would be a more
reliable guide. Present pOpulation and industrial expansion is
forcing foresters to produce more timber from less land. As a
result, forest managers need more accurate and reliable estimates
of the land's productive capacity. It is the purpose of this
thesis to investigate site factors associated with the causes of
the variability in site index of red pine plantations growing on
Rubicon sand.

The variability in the site index of red pine growing on
Rubicon sand could be the result of many factors. These factors,

for the purposes of this study, can be divided into two groups:

 

*Personal comrunication with Mr. William B. Botti, Staff Forester,
Forest Cultivation. Michigan Department of Natural Resources.

1

non-soil and soil factors. Of the two groups the non-soil factors
are the most difficult to measure, because of their complex nature,
or the time necessary to measure them. These factors include
climate, planting stock quality, genetic variation, handling of
the seedlings during planting, etc. Another factor, which is not
inherent to the site, is the site index curves used to determine
the site index values. Inaccuracies in the site index curves,
eSpecially on the younger or older portions of the curves, result
because the growth pattern of the individual trees in question

are not similar. Soil factors, by comparison, are relatively

easy to measure. Because soils are stable site features, they can
be quantified and described in detail. There can be little doubt
that as the medium for root growth, the soil is perhaps the most
important site factor. An understanding of the effects which the
above site factors have on site productivity is an imperative

prerequisite to intensive forest management.

BACKGROUND

Because soil scientists and foresters have technical jargon
peculiar to their respective fields it is often difficult for
them to communicate to one another. In an attempt to bridge this
gap I would like to clarify some concepts which need to be under-
stood by both professions viz., the soil scientists'system of

classification and the foresters concept of site.

§gil_Classification System
The current soil classification system in the United States,
also referred to as, Soil Taxonomy, is a classification according
to properties of natural soil bodies. The soil taxonomy is designed
to bring out the relationships of these natural bodies to their
natural and cultural environment (Soil Survey Staff, 1975). The
six categories used in the system in decreasing rank and increasing
number of differntiae and classes, are: order, suborder, great
group, subgroup, family, and series. The soil phases are further
subdivisions of any category, not in the system, but even phases
of series are important to land use (Thomas and Burroughs, 1973).
The Soil Taxonomy (Soil Survey Staff, 1975) defines the cate-
gories as follows: the higest category, ggggg, is divided into
10 classes based on the presence or absence of diagnostic horizons

or features resulting from the combined influence of the soil

3

forming factors. These factors are climate, and living organisms
acting on parent material over time as conditioned by relief. The
next category, suborder, describes properties such as the degree
of wetness or dryness, and/or the mineralogy of the soil. Next,
the ggg§§_ggoups describe the soil horizons, the moisture and
temperature regimes, and the similarities in base status. The
next category, subggoup, is divided into three groups of classes:
(1) typic, the central concept of the great group, (2) intergrades,
or transitional forms to other orders, suborders, or great groups,
(3) extra-grades, i.e. soils that have some properties not repre-
sentative of the great group and that do not indicate transitions
to any other known kind of soil. Next, families, groups soils
which have similar physical and chemical pr0perties and differences
of climate that affect their reaponse to management and manipu-
lation fbr use. Series, the lowest category, uses differentiae
which are mostly the same as those used in the higher categories,
but the ranges permitted in the prOperties are less than is
permitted in the higher categories. There are numerous additional
differentiae used to define a series that are not used in the
higher categories. Series criteria are closely allied to land
use interpretations. The soil phase, although not a category in
the soil taxonomy, is a further subdivision of a class in any
category used fer management and interpretive purposes. For
example: Rubicon sand, rolling,is a phase of Rubicon.

To illustrate how the soil classification works,the classifi-
cation of Rubicon sand, the soil used in this study, is described

in detail. Rubicon is officially classified as: sandy, mixed,

frigid, Entic Haplorthod.
Order: Spodosol.
The order is indicated by the last two letters of the name.
(Haplorthgg = Spggpsol). Spodosols are mineral soils which
lcontain a spodic horizon, i.e. an horizon in which amorphous
mixtwes of organic matter and aluminum, with or without iron,
have accumulated.
Suborder: Orthods.
Orthods are more or less freely drained Spodosols that have
an horizon of accumulation of Al, Fe, and organic carbon in
which no one of these elements dominates. All suborders
have two syllable names.
Great Group: Haplorthods.
These are more or less freely drained Orthods of midlatitudes
that have an albic and spodic horizon or, commonly, have only
a spodic horizon below an Ap horizon (plowed layer). The
spodic horizon may rest on a lower sequum that has an argillic
horizon (layer of clay accumulation), or on relatively un-
consolidated materials, or on rock. All great groups have
three or more syllables in their name.
Subgroup: Entic Haplorthods .
These are soils which:
a) Do not have argillic horizons below the Spodic horizon.
b) Do not have mottles in the spodic horizon, due to
water table fluctuations.
c) Do not have an horizon 15cm thick (6"), below the

spodic horizon or within lm (uo") of the surface,

which has a brittle matrix.

d) Do not have a lithic contact within 50cm (20") of
the surface.

e) Do not have a black intermittent upper subhorizon
that has a ratio of free iron (elemental) to carbon
that is less than 0.2.

f) Have less than 6% organic carbon in the upper 10cm
(H") of the Spodic horizon.

g) If a plow layer exists and extends into the Spodic
horizon it must have at least 1.2%Imore organic
carbon than the spodic horizon.

In parts of the United States where mean annual soil temper-

ature regimes are frigid, less than 8‘0 (H7’F), the albic

horizon (light colored horizon) of these soils commonly is

prominent. All subgroups have at least two capitalized

words in their names.

Family: sandy, mixed, frigid, Entic Haplorthod.

sandy - implies that the texture of the fine earth is sand,
loamy sand or coarser, but not loamy very fine sand
or very fine sand or finer; rock fragments make up
less than 35% by volume.

mixed - implies that there is less than 00%.of any mineral
other than quartz or feldspars.

frigid - implies (1) that the difference between the mean
winter and summer temperatures at a depth of 50cm
(20") is greater than 5°C (9’F) and, (2) the mean

annual soil temperature is less than 8‘C (u7’r).

Series: Rubicon.
Fer the official series description see Appendix I. This

outlines the specific, narrow range of soil prOperties
characteristic of this class in the lowest category of Soil
Taxonomy. The name is from a geographic feature near the
place it was first recognized.

Type: hibicon sand.
This is a kind of phase of the soil series, and implies that

the plow or surface layer is sand.

In Michigan, as an aid to interpreting the soils information
for land use purposes, similar soil series are grouped into what
are known as soil management grggpg: These groups are based on
soil properties to depths of 150cm (5 feet), and are designated
by a number and letter combination. The number indicates the
Profile texture and the letter indicates the natural drainage
Class. The numbers range from 0 to 5, where 0 represents the
fine clays (more than 60% clay) and 5 represents the sands. The
letflters are "a" for well drained, "b" for imperfectly drained,
and. "c" for poorly drained. Some soils are composed of strata
Wiiflu contrasting textures. In this situation a fraction instead
Of? a whole number is used. For example, a loamy sand material
50 - 100cm (20 - '40 inches) thick over clay loam to loam, on a
well drained site would be classified as II/ZII. For the sandy
Soils a decimal system is used to indicate the degree of profile

deve10pment. The numbers range from 5.0 to 5.7, where 5.0
represents a sand with a well developed profile and 5.7 differs

Iw'having a very poorly developed soil profile. Rubicon sand

is in the soil management group 5.3a.

The above infbrmation, plus Rubicon's suitability for timber
production, can be found in any soil survey in Michigan where

the Rubicon series has been mapped.

Site,

Site in its narrowest sense is that area where a stand, or
group of trees, is growing. It has been defined in many ways by
many researchers working with site and site quality. Heiberg
and White (1956) credit Tansley as proposing an acceptable concept
of site, "the sum of the effective conditions under which the plant,
or plant community lives." Tansley, an ecologist, used the term
"habitat" which is equivalent to the foresters term "site".

Rowe (1953) in describing the more dynamic and holistic concept

of site has found it to be "the complex interrelation and inter-
action of all features, inorganic and organic, past and present,
Which have resulted in the given fbrest stand". This idea paral-
1313 closely the ecologists idea of ecosystem. The forest scientist,
hOwever, is primarily concerned with the forest tree segment of

151W: ecosystem (Spurr, 19614).

Site has been defined by many others, e.g. Carmean, 1975;
Thomas and Burroughs, 1973; Husch, et al., 1972; Coile, 1952;
HiJJLs, 1952; etc. From their definitions two senses of site
eV01ve: (l) as an area for supporting tree growth and, (2) as
fhe capacity of that land to produce a forest stand. The latter
8ense, or concept of site, is termed site quality. Absolute

Site quality is theoretically measured by the maximum amount of

 

wood produced upon an area by forest trees (Spurr, 1952). It is

this concept of site, that of productivity, which is of primary

concern to the forest manager.

rm...

A's-1‘,-

 

LITERATURE REVIEW

Inmrtgnce of Site Quality

A survey of industrial forest managers by DeBell et a1.

(1977) emphasizes the need for intensive forest management. In

summary their survey showed that "use of most cultural practices

is increasing markedly, and that anticipated investments in

such practices are estimated to increase annual harvests through

1985 by 1'4 percent above the 1970 level...." It is inportant

then, for intensive forest management, to be able to accurately

estimate the potential of an area to grow trees as well as its

relationship to the management of the trees to be planted. The

Productivity of a site for tree growth is usually evaluated on

a stand basis. Considered in this way, site quality expresses

the average productivity of a designated land area for growing

forest trees (Husch, Miller, and Beers, 1972). The relationship

of site quality to the growth of forest trees is a difficult one

to measure. The factors of the site and the plants themselves

are interacting and interdependent, making it difficult to assign

Single cause and effect relationships. What then, should or

can be measured? In general, site quality can be evaluated in
two ways: (1) by measuring one or more individual site factors
which are considered closely associated with tree growth and/or,

10

 

ll

(2) by measuring some characteristics of the trees or lesser
vegetation considered sensitive to the sum total of the individual
site factors (Spurr, 1952; Husch, Miller,and Beers, 1972).
Hills (1952) cautions that the site is not merely the sum

of its parts, rather, it is "a whole which is something more

than the sum of its parts." In his holistic approach to the
evaluation of site, Hills integrates the complexities of
"climate, relief, geological materials, soil profile, ground

water, and communities of plants, animals and man".

 

Carmean (1975) believes that the holistic approach should
be used with caution because an integration of the various factors
of the environment and vegetation at each level of classification
nmkes the system difficult to comprehend. Arbitrarily rating
climate, moisture and nutrients is questionable; as a result,
the various "moisture regimes" are not well defined by means of
Standard quantitative soil moisture methods.

Circumventing these inherent problems of the holistic approach
(”flier researchers have elected to examine site quality indirectly
bY'Ineasuring the sites'observable characteristics and their effect
on wood production, and/or directly by measuring the total
quantity of wood produced as a reflection of the site factors.
(kflnprehensive reviews of their site research were published by
(3311e (1952) and more recently by Carmean (1975). In summary,

i“direct factors affecting site quality are: climate, t0pography,
landform type, geographical location, activities of man and other
Organisms, lesser and competing vegetation, quality of seed or

nursery stock and soils. Of the above factors, soils have been

12

the most extensively studied because they are the medium on which
the trees grow. Its pronerties are quantitatively defineablc

and are relatively stable. The relationship of the other factors
such as climate, seed or nursery stock quality, etc., are either
too difficult to measure, too unstable or are so general in
(nature as to have little practical significance. Direct measures FI‘E
of site quality are: vegetative growth, growth patterns and
associations, the volume of wood produced, and height growth at

a given age as an index of site quality. Recently, site research

 

has been facilitated by the use of computers and hOpefully in Lt'r
the future the more complicated relationships of site quality
will be understood.

Wambach and Lundgren (1965) recognize site quality as the
basic silvicultural variable. They credit Davis with the
following:

The forest manager should be site-conscious....Site

quality has a profound effect on the volume, value

and species of timber that can be best grown on an

area. It affects regeneration and cultural practices

such as cleaning, thinnings, prunings, and improve-

ment cuttings. Management practices should be

related to site.

Perhaps site quality can best be thought of in terms of
e¢3<Z>nomics. Forest management for timber production in the United
States is profit oriented and the ultimate goal of decision making
is to use the land as effectively as possible to produce maximum
eConomic return (Ralston, 1958; Mader, 1968).

Establishing a site's quality indicates to the forest manager

What to expect from the land. For example, Cooley (1970) found

that cone production and growth were increased more by heavy

13

thinning than by fertilizing in 53- to 55-year-old natural red
pine stands growing on medium sites and in a 20-year old stand
on a good site. Mason and Tigner (1972) studied an outbreak of
lodgepole needle miner in central Oregon and found that the

degree of infestation is influenced by a combination of environ-

 

mental and physiological factors that vary significantly under I w
1
different forest-site conditions. J
Total Site Factors Affecting Site Quality
Site in itself is not just one factor nor is it the sum of
L... _

 

all factors; rather, it is the sum of all the effective factors
(Heiberg and White, 1956). There are three basic methods used

to evaluate site. One method uses the volume of wood produced,
another uses the height growth at some specified age, and lastly,
one based on plant indicators.

The mean annual yield fbr normally stocked stands at the
Onlndnation of growth has long been considered the ideal measure
0f site quality (Spurr, 1952). In the site study of red pine
Plantations initiated by Mader and Owen (1961), of the Department
of? Forestry and Wildlife Management at the University of Massa-
CTHJsetts, an attempt was made to consider different aspects of
83kmwth to determine if the correlations with site factors differed
depending on the growth measure used. Better correlations
ochrred using periodic volume growth estimates than with total
height at 25 years, periodic height growth, or volume at 25 years.

In later studies, Mader (1963 and 1968), again found that cubic
feat volume growth curves appear to be one of the most useful

neans for making growth comparisons.

ll}

The basic assumptions made by those who advocate the use of

measured volume production as the most exact means of site eval-
uation differ in several respects from those who use height growth.

The major assumption is that height growth and volume growth are

not necessarily directly related. Another assumption is that

when differences in volume production measured in well stocked

 

stands do not agree with differences in observed height growth, 3
these variations are not caused by density factors or lack of i
full production on the site but are, in fact, real differences
in Productivity, and therefore height growth is basically unable

to fully reflect differences in volume growth potential (Mader,
1963).

According to Carmean (1975), when adequate yield information
is lacking, as is the case for many species in the United States,

Site index has been used to considerable advantage as an indicator

of the forest land capability. Site index is sinply the height

of the dominant and codominant trees at a reference age usually

50 or 100 years. The height of free-grown trees of a given

species and of a given age is more closely related to the capacity
of a given site to produce wood of that species than is any other
one measure. The height growth of the dominant and codominant

trees has usually been taken as representative of stand height

£01? site index work. Because site index is so important for both

directly and indirectly estimating site quality, forest researchers
and forest managers should understand that even when suitable
dominant and codominant trees are available, the accuracy of

site index estimates may be affected by several stand and tree

15

conditions. Furthermore, an understanding of the methods used

for constructing site index curves is important. The reason is

that the kind of data and the computation methods used will

determine the accuracy of site index curves, and thus, determine

the precision of the site index estimations (Husch, 1963; Spurr,

195a; Carmean, 1975). m“

Site index is not a foolproof method and its weaknesses are

well known. First and foremost, the technique is sound only if

the average site quality is the same for each age class. Second, p

 

'51

it is assumed that the shape of the height-growth curve is the
Same for all sites. Third, it cannot be safely used on stands

less than 15 to 20 years old. Fourth, the terms dominant and

COdominant are subjective and two foresters may differ widely
in their concept of what constitutes dominant and codominant

trees. Fifth, it is difficult to see the tops of trees in a tall

and dense stand and it is therefore, difficult to measure their
heZlghts accurately. Finally, silvicultural practices such as
tlli‘nning may also change the average height of the dominant and
Qodominant trees without, of course, changing the actual site
quality (Spurr, 1952 and 1955; Wakeley and Marrero, 1958).
Stem analysis is another method of arriving at site index.
Simply, it is the felling and sectioning of a tree; counting the
annual growth rings and measuring the diameter at each cut surface.

PI‘om this information the past develOpment of a tree's height and

age can be determined. This method too, is not without its

First of all, it is the means of determining the
Secondly,

weaknesses .

past growth of individual trees and not a forest stand.

16

tree rings represent springwood and summerwood which are presumed
to be produced every year, but this is not always the case (Spurr,
1952). Also, the true relationship of height and age may not be
accurately represented if the discs are taken at a uniform
distance apart because the height from which the disc was taken

will frequently not coincide with the total height of the tree

at any given age.

Many stands suitable for site index measurements may not

contain the tree species for which the site estimates are desired.

 

Suitable dominant and codominant trees of several species may be
present, but no useable trees of the particular desired species

may occur. For such stands we can use the tree Species actually

Present for estimating site index. Species comparison graphs

can be used to convert the site index of the species present to

the site index of the desired species. Comparison graphs and site

index ratios have been prepared for several forest species in
Various parts of the United States (Doolittle, 1958; Della-Bianca
and Olson, 1959; Carmean and Vasilevsky, 1971; Carmean, 1975).

For tree Species that have limbs showing distinct annual
WI“lorls the cumulative length of three to five internodes beginning
at: breast height has been suggested as a measure of site quality

DE11:“ticularly for young stands (Ferree et al., 1958; Wakeley and

MaI‘rero, 1958; Day et al., 1960). This method has many advantages:

(1) it does not require curves of height over age for direct
Q0triparison of sites in plantations of different ages; (2) it is
not affected by the tree's establishment period; (3) total age

or total height need not be known, therefore errors due to their

17

measurement can be eliminated; and (II) it can be measured

easily and rapidly (Wakeley and Marrero, 1958). Disadvantages

of the method include the effects of short term climatic fluctu-

 

ations, and the fact that sometimes the early growth of a stand

does not accurately reflect later growth particularly when 80

to 100 year rotations are used (Alban, 1972; Carmean, 1975).

Trees as well as the lesser plants comprising the forest

understory are all part of the vegetation which can be used to

 

estimate site quality. For this discussion, vegetation will
L _.-

refer to the non-tree segment of the forest flora.

The use of indicator plants, or phytometers, as a means of
estimating site quality is based on the theory that certain
key species in the forest reflect the overall quality of the

Site (Husch, 1953). Although the understory'plants are apt

to be influenced by stand density, past history, and the composition

of the forest they have a narrow ecological amplitude and therefore,

are useful as Site indicators (Spurr, 1969). Husch, et a1. (1972)

point out that when using phytometers to estimate site quality
car-e should be taken to keep the relationship on broad terms as

this method was developed in boreal forests where the stands are

extensive and the tree species few.
Using vegetation as an indicator of Site quality has not been

widely accepted in the United States (Carmean, 1975). Several

criticisms have been made of this method which may partially

eitplain why it has not been widely applied in the United States

(Coile, 1938): (1) many site types are closely related to features

of geology, soil and t0pography thus land classifications could

 

18

be based on these features alone rather than on understory

(2) on Similar soils the kind of overstory tree may

(3) trees

vegetation;
have a pronounced affect on the understory plants;

often have deeper root systems than the understory plants thus

deeper soil horizons may affect the tree and have little or no

affect on the indicator plant; (ll) stand density affects the I?"
understory plants by controlling the amount of sunlight which I
reaches the forest floor; and (5) many indicator plants are not
evident during the dormant seasons and thus cannot be used.

r

 

Mensurational methods have been proposed for uneven-aged stands

and for stands that lack trees suitable for directly estimating

site index using conventional methods. For example, Gevorkiantz

and Scholz (1999) in studying the oak forests of the Upper
Mississippi Valley found that the acceptable site index relation
between age and height of the dominant trees - was not applicable.
They showed that the product of the average basal area and average

helight of the dominant trees when used as an index gave more
meliable results, especially for stands which were below normal
S":Ocking.

Using the soil series as a method of estimating site quality

has met with limited success. Shetron (1969) has found that

f0lt‘est managers would be able to estimate site quality for jack
Pine on certain soils (Grayling, Graycalm, Montcalm, and Deer
Park) whereas on other soils (AuGreS, Croswell, and Rubicon)

they could not. For the later soils, he suggests refinements

be made, especially in depth to mottling in the AuGreS and

Croswell soils .

19

van Eck and Whiteside (1963) and Shetron (1969) have found
that site index is greatest on well drained management groups

3.0a and lI.0a but it decreased as the soil texture became finer

or coarser. The data presented by Shetron Showed that the site

indexes of jack pine, sugar maple, red oak and big tooth aspen

all followed this general trend.
Studying red pine in northern lower Michigan, van Eck (1958)

found that site indexes varied little over a wide range of

 

50' T (‘3‘

moderately coarse to fine soil profile textures, but showed

considerable variation between soil series within the coarser

textured groups. Studying other tree Species in the same area,

Shetron (1969) found that the soil series of the coarser textured

groups only accounted for part of this variability.

W931 gie {suitors Aifecting Site @ality

Another approach to evaluating Site quality is to examine one

or a few factors considered to be important. Much work has been

done in this area and many reviews of the literature, sunmarizing

the results, exist. A review of the early literature was presented

by Coile (1952) and, of the more recent literature by Carmean (1975).

Many factors of the site such as soil, climate, geographic

1°teation, altitude, stock quality, and others affect site quality.

Of all the factors, soil has received the most attention because

it is the principle medium in which the trees grow and its
PPOPerties are relatively easier to examine and quantify (Locke,

1991) . Mader (1961) adds that the "understanding of the influence

0f soils on growth rates and patterns of forest stands is essential

to obtain maximum production from forest land." Coile (1952)

um - "7 "

20

however, points out that if "all forest land were covered with
well-stocked stands of sufficient age for the entire solum and
upper stratum to have affected their growth, there would be
little practical need for studying the relation between soil
properties and growth because the volume of wood per acre at a
given age would be a direct measure of productivity..." Since 'rnfih
most forests are not of such stocking or age the obvious need i

to evaluate site factors arises. Which site factors are most

important depends on the area and species in question. In

 

general, it can be said that factors which affect the soil moisture
arui nutrients, and the tapography are the most important.

The effect of soil moisture on Site quality cannot be over
emphasized. White (1958) in his discussion of available water
States that, "except in situations of acute nutrient deficiency,

a scheme which would estimate the amount and distribution of
a‘Vtailable water during the growing season would probably most
acQ‘urately evaluate site."

Soil texture is extremely important in affecting Site quality
beClause it has a major influence on the soil moisture, soil
‘3}lfiandcal, and soil air relationships to root development (Heinselman
and Zasada, 1955; Spurr, 1969).

The effect of water table depth has been investigated by
‘nemry researchers and found to correlate well with tree growth

(Wilde and Pronin, 1999; Mader, 1968; Page, 1976). Studying the
depletion of subsoil moisture by apple orchards in NebraSka,
Wiggans (1969) estimated that in 17 years the trees utilized

70 to 75 inches (178 to 190cm) more water than they were able to

21

secure from annual precipitation. The additional water was

removed from the soil to depths of 25 to 30 feet (7.6 to 9.1m).

 

Related to ground water depth is the soil drainage class.
Barrett and Goldsmith (1973) found that the most important

factor in predicting white pine growth was the amount of avail-
Mader and Owen

D

4.1—:

able moisture as measured by drainage class.
(1961) found Similar results for red pine in Massachusetts.

Any characteristic of the soil which affects water movement

 

‘m...iu l. ‘

through the profile also has a pronounced effect on site quality.
Hannah and Zahner (1970) found that the site index for natural
jack pine and big tooth aspen stands, and stem wood production
in red pine plantations are Significantly higher on soils with
Prominent texture bands, whether pedogenetic or nonpedogenetic,
than on soils where bands are absent or are weakly developed.
White and Wood (1958) also studying red pine plantations found

marked variation in tree growth as a result of a fine soil layer.

Mth better growth resulted when the fine soil layer was six

f ee1: or less in depth. In Saskatchewan, Canada, Rowe (1953)

folmd that a frozen soil layer was able to affect soil drainage
by creating a perched water table as late as July.
Other factors have been found to affect soil moisture to

valflying degrees. Some of the more inportant factors are: rain-
fall, organic matter, Ap horizons as they affect organic matter,
and free iron and organic carbon (Minckler, 19lI3; Wilde and Pronin,

1919: Heinselman and Zasada, 1955; Mader and Owen, 1961; Mader,

1968; Shetron, 1979; Carmean, 1975; Page, 1976).

 

22

Much of the work done with soil nutrients has either been
done in conjunction with soil moisture or as a result of fertili-
zation. very little has been done with the naturally available
soil nutrients themselves. In a qualitative manner Coile (1952)
described soil nutrients as "perhaps a limiting factor in
forest growth on deep, excessively drained, siliceous sands in
humid climates." This is in harmony with Heinselman and Zasada
(1955) who concluded that aspen Sites are apt to be poorest on

sandy soils, where nutrient levels are low, and where moisture-

 

rmataining capacity is also low. They also state that soil :-~

tweaction, or pH, is probably not a significant site factor for

aspen.

More recently, Page (1976) studied the relationships between

Site index of black spruce and balsam fir, and soil and topo-
g'I‘aphic characteristics in two areas of Newfoundland. After
examining 103 variables from 300 sample plots he found that

'50:) one site factor was sufficiently closely related to site
itlélex for prediction purposes; in most cases six- or eight-
"E¥t¥iable equations were necessary to account for more than 60
1’5313cent of the observed variation." He also found that the soil
nu1:rient status was not as important as soil moisture but it was
'“‘31?e important than effects of t0pography. Of the soil chemical
Pr0perties he tested, the following were found to be the most

important (in order of decreasing importance): total nitrogen

(percent by volume) at 2.5cm (1") depth; C/N ratio at 30cm (1 ft.);

total N (percent by volume) at 30cm (1 ft.); and pH at 15cm (6").

 

23

When he compared measured site index values to predicted values
he found no significant differences.

Topography also affects site quality to a large extent
because it influences the 5011's physical and chemical properties
(Spurr, 1959).

In Newfbundland, Page (1976) found slope percent and the
change in elevation to be the most important topographic factors
affecting the site quality for spruce and fir. This agrees with
many other studies that have found the following tOpographic
factors to be significant (Carmean, 1975); slope position, degree
of slape, sloPe exposure, elevation, distance to bog, and
latitude.

The geomorphology and geology of an area is also important.
Lutz (1958) points out that landforms and rocks are intimately
related to parent material of the soil and therefore indirectly
affixns its texture, structure, and nutrient status. Finally,
once the factors affecting site quality have been established
they must be organized and presented so that they will be utilized
to their maximum benefit. The soil survey accomplishes this end.
Summarizing the soil surveying methods used by the Weyerhaeuser
company in the Pacific Northwest Steinbrenner (1973) says, "the
soil survey provides information that is basic to sound forest
,management and its use can only increase as more interpretive
detail is developed." Retzer (1958) discussing soil as a factor
affecting forest vegetation states that "soil taxonomic units
represent the most effective way of stratifying significant

physical and natural differences in the landscape."

STUDY AREA AND PROCEDURES

A) Nature of the Study Area

The study area was confined to two counties, Grand Traverse
and Wexford, in the northwestern part of the lower peninsula of
Michigan, More specifically, 3 plots were located in the south-
eastern corner of Grand Traverse county and 7 plots were located
in the eastern half of Wexford county (See Figure l). Exact
locations and descriptions of each plot can be found in Appendix
IV.

Geology

The last ice sheet of the Wisconsinan ice age, or glacial
period, formed the surface features of these two counties.
Between 23,000 and 12,500 years ago, when this ice sheet melted,
it left behind a series of moraines, outwash plains, lakes, and
beds of former lakes, and abandoned drainage Spillways. Part of
one of the most extensive moraine systems in Michigan called,
the Port Huron moraine, extends across Grand Traverse county in
an east-west direction forming the northern boundary of the lower
third of the county. South of this moraine, in the southeast
part of the county, is a sandy outwash plain on a high plateau.
To the south of this outwash plain in Wexford county, there are
other moraines which are bounded by other outwash or till plains
‘that have been dissected by glacial Spillways. A conspicuous

2H

25

Moraine

E]
Lake Bed I
Plain

Sandy Outwash Plain

GRAND TRAVERSE

COUNTY

/Port Huron

Moraine

 

WEXFORD COUNTY

 

 

 

 

Figure 1. Outline of the Two County Study Area, its Surface
Geology and Plot Locations.

* Plot Location

26

geomorphic feature in Wexford county is a former glacial lake
in the southeast portion of the county. In general, outwash
plains, which cover extensive areas in both counties, are respon-
sible for the prodigious amount of sandy soils found in the study

area .

Climate

The climate of the study area is humid, cool, temperate and
continental. The effect of lake Michigan is minimal and the
climate, therefore, is considered uniform over the entire area.

The average annual precipitation from 1936 to 1973 was 31.”
inches (798mm). One year in 10 will receive less than 28 inches
(711mm) and one year in 10 will receive more than 37 inches
(990mm). The average monthly precipitation throughout the year
ranges from 1.5 inches (38m) in February to 3.9 inches (86mm)
in September (National Climatic Center, no date; Weber, et al.,
1966).

The average annual temperature from 1931 to 1973 was 92.6 ‘1‘
(5.9‘C). During the year the average monthly temperature ranged
from 18‘F (-7.8’C) in February and 67'F (19.9 'C) in July. During
the winter, average snow fall is 70 to 80 inches (178 to 203cm)
(National Climatic Center, no date; Weber et al., 1966).

In general, the climate of the area is characterized by a
wide range in temperature between the extremes of summer and
winter plus an irregularly distributed and relatively abundant

rainfall (Shetron, 1969).

27

Vegetation

Much of the area today is either being farmed or is forested.
According to Weber et al., the trees of the area can be placed
into 3 groups: (1) sugar maple (Acer saccharum), beech (Egggg
grandifolia), elm (Ulmus americana), and other hardwoods on the
loams, sands with well developed spodic B's, and more fertile
soils; (2) white pine (Pinus strobus) and red pine (Pinus resinosa)
on the sandier, less fertile soils; and (3) white cedar (Ihgga
occidentalis), balsam fir (Abies balsamea), and black Spruce
(Picea mariana) in the swamps (Weber et al., 1966).

Less common trees are: black oak, white oak, quaking aspen,
big tooth aSpen, balsam poplar, ironwood, yellow birch, paper
birch, black cherry, white ash, black ash, basswood, jack pine,
tamarack, hemlock, and juniper (Weber et al., 1966).

After the original timber was harvested, large areas were
burned; then, through natural seeding, were covered by stands
of aspen, oak, pin cherry, and other trees.

The ground cover in the wooded areas consists of bracken
fern, sweetfern, dogwood, sumac, and many others. Blueberries,
bladk berries, raspberries, and strawberries grow in cutover

areas or swamps that have not been burned (Weber, et al., 1966).

Soils
The soil taxonomy (Soil Survey Staff, 1975) defines soils as:

the collection of natural bodies on the earth's surface, in
places modified or even made by non of earthy materials,
containing living matter and supporting or capable of
supporting plants out-of-doors. Its upper limit is air

or shallow water. At its margins it grades to deep water
or to barren areas of rock or ice. Its lower limit...

(is) ... to the not-soil beneath....

28

This concept of soil will be used for the remainder of this thesis.

Soil pr0perties are the result of: (l) the parent material;
(2) the climate; (3) organisms, both plants and animals, which
live on or in the soil; (H) the t0pography; and (5) the length of
time which has elapsed. Since this study involves a relatively
small geographical area the above mentioned soil formation factors
are considered uniform throughout the area. Their inter-relation-
ships are complex and involved, and are beyond the scope of this
paper. A.more detailed discussion of these factors is presented
by Buol, Hole and McCracken (1973), and by Jenny (1991).

Rubicon is the only soil series reported on in this study.
The official series description of Rubicon can be fbund in Appendix
I. (The series description for Graycalm is also in Appendix I fbr
comparison). Some general comments on Rubicon follow.

The Rubicon series consists of deep, well drained soils
formed in sandy glacial or glacio-fluvial deposits. The solum,
including the A.and B horizons, is generally 20 to 50 inches
(50 to 130cm) thick. The degree of spodic horizon development is
between that of Kalkaska and Grayling sands. Kalkaska has a dark
upper spodic horizon more than 3 inches (7.6cm0 thick and Grayling
has little or no A2 or spodic horizon development. On poorer
drained sites Rubicon is typically associated with Croswell, AuGres,
and Roscommon soils. On well drained sites it is usually associated

with Kalkaska, Grayling, and Montcalm soils.

B) Selection of Sites
To save a prodigious amount of field work it was necessary

to obtain available forest cover type and soils information from

29

both the United States Forest Service and the Michigan Department

of Natural Resources prior to going into the field. On the basis

of this information possible study areas were located. These

areas were then examined in the field and, if they met the

following criteria, were established as temporary plots:

1)

2)

3)

Plantations had to be 20 years of age or older from seed
and even aged.

Size of the plantations had to be large enough so that
trees in a 1/10 acre plot were not influenced by edge
affects.

The soil had to be Rubicon sand and be at least 1/10
acre in extent. This was determined on the basis of

5 auger borings to a depth of 5 feet (1.5m). If the first
boring proved to be Rubicon sand it then became the plot
center. Four additional borings were then made at the
edge of the plot, one in each of the four cardinal
directions (N ,E ,S ,W) .

If the above criteria were met, the area was then established as

a temporary plot, and the next step was to collect the data.

C) Field Procedures

1)

A soil pit was dug to a depth of 6 feet (1.8m) at the plot
center, and the soil profile was deScribed in the standard
manner. The profile was then checked further to a depth
of 10 feet (3m) or more with an auger. See Appendix IV
for the descriptions and laboratory data and notes at

each site.

30

2) Soil samples, of approximately 1 pound (SOOgms) each,
were taken from each soil horizon in the upper 5 feet
(1.5m) of the profile.

3) Basal area of the plot was measured from plot center with
a 10 factor prism.

H) Five dominant and codominant trees nearest to the soil
pit were felled and their heights measured to the nearest
inch.

5) Each tree was sectioned into 9 foot (1.2m) lengths starting
at one foot (.3m) above ground, with a one inch disc
taken from the upper end of each 9 foot section.

6) The percent lepe, aspect, plot position, and type of

landform were recorded.

D) ‘Laboratorngrocedures

1) Soil textures were determined for each sample using dry
sieve analysis techniques.

2) Soil samples were analyzed for the presence of P, K, Ca,
Mg, and the soil reaction (pH), and organic carbon accord-
ing to soil testing procedures for available nutrients,
at the Michigan State University Soil Testing Laboratory.

3) The annual growth rings were counted and diameters

inside bark were measured for each tree section.

B) Computer Analyses
Site index curves were developed from the stem analysis data
by generating an equation which best explained the relationship

of tree height to age. From this equation a family of curves

31

were also derived (site index curves). This procedure in detail is:

I)
2)

3)

The tree data from all plots were pooled.

Regression analysis, using the least squares method, was
employed to generate several equations to represent the
relationship of tree height and age. (Note; in using this
method, Freeze (1969) states that the following assumptions
must be made: (1) the data is from a p0pulation for

which the variance is homogeneous i.e., the variance of

the Y values (height) about the regression surface (re-
gression line) is the same at all points or with all combi-
nations of X values (age); (2) for the sample units the
deviations of the Y values from the regression surface
must be independent of each other, i.e., size and direction
(+ or -) of the error for one unit should have no rela-
tionship to the size and direction of the error for any
other unit in the sample, beyond the fact that they are
from the same population; and (3) the x values are measured
with essentially no error).

Covariance analysis was used to determine whether separate
prediction equations should be used for each plot or
whether all of the plots should be represented by a

single equation. This procedure involves comparing the
levels and slapes of the equations for each plot with

the mean equation fbr all plots. This step was done

mathematically and visually.

32

a) Mathematically the equations were tested as follows:

df, Rss MS

 

Total of individual plots (i) dfi Rssi MSi
Total (all plots pooled) (t) -dft —Rsst -MSt
Difference (d) dfd Rssd MSd
F = MSd/MSi

where;

df 3 Degrees of freedom
Rss 8 Residual sums of squares
MS 8 Mean square
M51 = Rssr/dfi
MSd = Rssd/dfd
If the computed P value was larger than the value
of F, at the 5% level, in standard F tables, the
individual equation should be used to represent the
data because the individual equation has a siginif-
icantly better fit.
b) The equations were tested visually by comparing
the graphs of the individual plots to the graph
of the mean curve. No:curves were extrapolated
beyond 5 years for which the data was available.
(The above steps a and b are summarized in Appendix
II, Tables C and E).
H) Each equation was subjected to analysis of variance.
If the equation was not significant at the 5% level of
probability it was rejected.

33

5) The equations were subject to a test of linearity to see
if a straight line or quadratic equation would best fit
the data. (This test is summarized in Appendix II,
Table B.)

6) The equation which best described the mean relationship
of height and age i.e., the equation with the largest
coefficient of determination (r2), was chosen to generate
the mean curve. (These equations and their r2 values
are summarized in Appendix II, Table A0.

7) The site index, that is, tree height at age 50 years,
for each plot was determined from both the harmonized
and the polymorphic curves developed from the data of
this study and compared with Site index curves developed
for other studies.

8) Stepwise regression analysis was employed to screen Site
factors which were believed to influence site index.

They are listed in Appendix III.

F) Velume Determination

The cubic fbot volume for each tree was determined for 5
year intervals, by measuring the radius of those discs which fell
at the end of a 16 foot (9.9m) log, i.e. discs cut at the heights
of l', 17', 33', and 99' (.3, 5.2, 10.0, 19.9m). The procedure
is as follows:

1) The total average radius was measured to the nearest 1/10

inch.
2) The radius was measured for each 5 year's growth counting

from the outside inward. (NB: The volumes calculated

3)

LI)

5)

5)

7)

8)

39

for each tree may not be completely accurate because
the discs were measured after considerable drying and
in some cases slight cracking had taken place.)
The radial growth was then plotted on 10 squares to the
inch graph paper. Height in feet was placed along the Y
axis and twice the radius squared, (2r)2, in square
inches was placed along the X axis.
To organize the data it was tabulated in the following
manner:
massagili

where,
TH 8 total height in feet at the age in question.
DH - disc height in feet. (The disc taken at 1 foot was

placed on the X axis where Y-O).
r a radius inside bark in inches.
2r - diameter inSide bark in inches.
A cOnnecting line was drawn between two adjacent points
such that a negative curve was obtained. A final connect-
ing line was also drawn from the last data point to total
tree height.
The total volume of the tree was calculated by multiplying
the area under the curve (in square inches) by .5959.
The volume fbr each of the trees growing on each plot
were averaged and the average volume was used to represent
the plot.
Regression analysis was employed to express the relation-

ship of volume and age in the same manner as was done fer

9)

10)
ll)

12)

35

height and age. A mean curve was developed by pooling
the data of all trees.

A family of proportional curves were developed from the
mean curve.

Vblume at age 50 was determined from the above graphs.
The volume at age 50 thus determined was converted to a
per acre basis by multiplying the volume per tree by 200
trees per acre. Two hundred trees per acre was used because
plot 3, cut at age 50, was typical of the plots used in
this study and contained 198 trees per acre.

Trees per acre was determined according to the following
formula (after Hanson, 1975): Basal area/.OOSHSH (Dia-

meter outside bark at breast height)2.

RESULTS AND DISCUSSION

The data from this study will be considered in two groups:

the tree and the soil data.

Tree data

The height over age data from stem analyses were plotted as
a scatter diagram in Figure 2. In order to group this data into
meaningful management units site index curves were developed.

The first step in developing site index curves was to
calculate the mean, or guiding, curve which best represents the
relationship of all height over age determinations using regression
analysis techniques. Two general types of linear equations were

tested: straight line, Y I a + bx, and curvilinear, Y I a + bx + cxz-

A.aummary of the equations tested and their r2 values is presented
in Table I.

The equation which best fit the data was of the curvilinear,
or quadratic, form:

Y I a 4- b(X1) + c(X2)

where,
Y I height
X1 I age

X2 I age squared or (X1)2

a, b, and c I constants

36

37

 

 

O
é .. Y = -9.017 + 1.399(x) + .00520(x)2
g. o
r2 = .9202 , 4
O. L- 0: 0+ .
8 0
9““: .i
O
903». 0+0 .«fiu’
2L...
0 90
ID 0
O we 0 99 N
u o O. O.
o g e- i’”””
8 * 0 0‘00 9 mo 90 o
v o
P O O M 6‘
g 9; '- 00.... O
O
a O 00 00 O
:l: ‘3 o o
‘ I—
a o
O M O
O O
0
‘5r-
.n
W
-_ w
C3 I l I J l l .1
a.T 10.0 20.0 30.0 40.0 50.0 60.0

AGE (years)

Figure 2. Scatter Diagram of Heights and Ages of All Red
Pine Trees Sampled on Rubicon Sand.

.— ~..u~.~.f~ ....u ptael.‘ -.:~..—.s>.:~ 1......- u:_:e..~ Lee \r.-~2h.:..f. .~ ..b~a.-.~.~.
g

38

m:.H

+l

Am.H

+|

om.:

+I

mo.m «

 

Anommv .m.m

m.mH mmonw.
:.NN cmmsc.
:.mH mmomm.
o.mH mHmHm.

.B No

eoeem oasossrm u mm
cowumwnm> mo pnowofimmwoo u >0

a~x\aanm.mH + Ax\uosmm.oa . maa.m u s nos 9 ~x\ao + x\ar + a u s
Axkvflmg - as; u s was u {as + a n s

Amxcommoo. . Axoasm.u + aao.a- u s u mxo + so + a u S
Axomoa.a + smo.a- u w u xo + a u s

 

powwow composum

 

.ow< one power:

some mo memo ooaoom on“ Eoem ooooam>on mnowumsvm mo humeesm .H manna

39

"a" was found to be equal to -H.017, "b" equal to 1.399,
and "c" equal to .00520 (Table l and Figure 2). The standard
error of the estimate was calculated to be 9.90 feet. A,correlad
tion analysis showed the multiple correlation coefficient (R) to
be .95929. The multiple correlation coefficient is a measure of
the degree of association of the estimated height with that of
age and age squared. The multiple correlation squared, or co-
efficient of determination, is equal to .9202. This means that
approximately 92% of the variation in height is associated with
age and age squared. An analysis of variance Showed the multiple
correlation coefficient to be highly significant.

It can be seen that the curvilinear regression is more
representative of tree height-age growth, and explains somewhat
more of the association of height and age then does the straight
line regression by comparing their coefficients of determination.
The curvilinear regression explains more than 92% of the relation-
ship while the straight line regression explains less than 92%.

Is this difference significant or is this relationship
adequately explained by the straight line regression alone? A
test of linearity showed that there is a significant curvilinearity
in the regression (Appendix II, Table B). Therefore, the addition
of the age squared term explained a significantly greater amount
of the variation between height and age. This equation is
graphed in Figure 2.

Because differences exist between plots, the question
arises as to whether separate prediction equations should be used

for each plot or could all of the plots be represented by a single

In

... -‘ __r

 

 

 

#0

equation? An analysis of covariance indicated that there is a
significant difference between the individual curves and the
overall curves (Appendix II, Table C). This means.that the data
should not be represented by one overall curve because the amount
of variation explained by the individual equations is signifi-
cantly greater than that explained by the overall equation. This
analysis tells us that in mathematical terms a significantly better
fit will be obtained if more than one equation is uSed to describe
the relationship of height and age. It does not tell us how many
equations are necessary, nor if the differences are important in
terms of current management objectives.

That more than one equation is necessary indicates that the
growth patterns or growth rates are different between plots and
polymorphic growth curves should be develOped. These growth
characteristics were checked visually and, in the case of growth
rates, also checked by comparing the tangents or skme’coefficients
"b" of the individual plot equations. Each plot curve was graphed
and compared to the mean, or overall, curve (Appendix II, Table B).
These curves show that there are two groups of plots. One group
of plots (3, 8, 9, and 10) growing faster then the mean curve of
all plots and another group (1, 5, 6, and 7) growing slower than
the mean curve. The two remaining plots (2 and 9) were growing
similar to the mean curve and might be placed in either group.
Regression equations were calculated for both groups. The best
fit was obtained with the equation, Y I a + bx + cxz, using plots
3, 8, 9, and 10 for the faster growing plots and plots 1, 2, 9, 5, 6,

and 7 fbr the slower growing plots (Table II). These equations

L ...... ...: Q~£rnh >Ltzht-ufl .NN .WNQQLH

Hl

 

 

 

 

 

 

 

a.ma owns. o.- seam. a.ru mass. A.aa mass. Aoosoamc ensnmna

A.au swam. a.a~ mass. ~.s mama. s.s comm. Assesses as a.r a m N
HH pom

a.aa some. a.sN seem. o.oa mmss. o.oa sass. Assesses a.o.m”snmna

a.oa mass. m.ma same. c.s ammo. e.s some. Assesses or a s m
H row

.>.o ms .>.u me .>.u me .>.o . Ne areas

 

n w mod

X\wn+o I.Nwoa

Nxo + xo + u I > xo + m n w

soaps

u

 

.muoam mo masons wousonw nozoam one noummm may mo mud
our sewer: or» so osoaosssm.osr sou earns anagram .HH oases

92

Y = -s.021 + 1.613(x) -.00565(x)2 Upper

 

 

r2 = .9836
Y = -l.936 + 1.009(x) +.000979(x)2 Lower
a r2 = .9933
‘ I—
2 .
O
I N
‘2 .."
8 "' Faster t .
\ 3
O 0* 00* ”9‘3"
0 O
8 L- o . N
A o o ' w ‘1 Slower
z o. _ ' .6 O.
o 2
fi 6 O N 6
~..I ,
F. c: ' °
II: fiél-I .
IE! .
- 9 so 9
III
I on O 0
° " I
N 'I ’
O O W
9: P- 0”“
W
__ M
C3 I I 11 I .I I I
o.' 10.0 20.0 90.0 49.0 60.0 60.0

AGE (years)

Figure 3. Mean Growth Curve for the Faster and Slower Growing Plots,

93

were than graphed as shown in Figure 3. Again, the question was
addressed as to whether separate prediction equations should be
used for each plot within each group or could each group be re-
presented by a single equation. AS before, analysis of covariance
showed that a significant difference existed between the individual
and overall equations for each group (Appendix II, Table D). This
means that the data of each group should not be represented by

one equation.

However, before the data was further subdivided an attempt
was made to see if the first subdivision of the data, that is,
dividing the plots into faster and slower growing groups, had
resulted in any inportant improvements over the mean equation in
terms of management. This was done by using the mean curves
(the mean curve fbr all 10 plots and the mean curves of each
group of faster and slower growing plots) as guiding curves for
developing proportional site index curves. Using the mean
equation for all 10 plots resulted in the standard harmonized
site index curves (Figure, 9), whereas, using the mean curves for
the two groups of plots resulted in the polymorphic site index
curves (Figure 5). The mean curve for the 9 faster growing plots
was used as the as the guiding curve for the higher site index
classes of 60 and 70 while the mean curve fer the 6 slower growing
plots was used as the guiding curve for the lower site index
classes of 90 and 50. The harmonized and polymorphic site index
curves were then superimposed (Figure 6).

For site index classes 60 and 70 there is very little difference

between the sets of curves. For site classes 90 and 50 some

HEIGHT (foot)

20-0 30.0 40.0 50.0 80.0 70.0

10.0

 

0'1

Figure 9.

99

 

l 1 J 1 1 I I
o.I 10.9 20.0 30.0 40.0 69.0 90.0

AGE (years)

Harmonized Site Index Curves Using the Mean Growth
Curve of All 10 Plots Combined as the Guiding Curve-

 

95

 

 

 

o
‘ Iv—
‘9
8
‘2
A
‘6 o
' h-
}! 13
v
SE :3
' r-
5 8
-
III
I ‘3
a _
u
9 .— Mean Curves
2 Faster plots __
Slower plots _ _ __
g-r—
‘3 L, I . I I I I I
0.r 10.0 20.0 00.0 40.0 50.0 80.0
AGE (years)
Figure 5. Polymorphic Site Index Curves. The Mean Growth Curve

of the Faster Growing Plots was Used as the (hiding
Curve for Site Index Classes 70 and 60. The Mean
Growth Curve of the Slower Growing Plots was Used as
the Guide (hirve for Site Index Classes 50 and 90.

 

4l\ Lrndc7hu:

ESQ?

HEIGHT (feet)

20.0

40.0 50.0 80.0 70.0

30-0

10~0

Figure 6 .

1+6

 

 

 

 

I.
/
h l
I—
— / Harmonized
/ Polymorphic __
I I I I I I I
OJ 10.0 20.0 30.0 40.0 50.0 80-0

AGE (years)

Comparison of Polymorphic and Harmonized Curves.

.. Cc
. . . . .g n .E S v
.‘v

1'
.n o as:

a? .
V‘

97

differences exist especially when the stands are less than 35

years old. These differences however, are only as much as 5 or 6
feet. After age 35, and when the index curves become more accurate,
the differences are less than 5 feet. Therefore, since the dif-
ference between the two sets of curves is always less than one

site class, little advantage is gained by developing polymorphic
site index curves.

As site quality is being meaSured in terms of tree growth, the
comparison of their optimal growth rates is justified (van Eek,
1958). The Optimal growth rate is equal to the SIOpe, or tangent,
of the regression equation and is referred to as the "b" coefficient.
Table III is a list of the "b" coefficients. The largest coefficient
is 1.955 and the smallest is 0.962. This means that the fastest
average growth rate was approximately 1.5 times as fast as the
slowest. However, 70% of the plots fall within the range of 1.103
to 1.372, which means that for most plots the fastest average rate
of tree growth is 1.2 times as fast as the slowest.

After observing the small differences between the two sets
of site index curves, and the small differences between the average
growth rates of each plot, it can be said that in terms of current
management objectives, no important advantages are gained when
more than one equation is used to describe the relationship of
height and age, Table IV (This Study). Further, since the first
subdivision of the data into faster and slower growing groups
was unimportant, any further subdivision of the data would not

bring any further advantages.

98

Table III. The "b" Coefficients, or Slape Constants, of the
Straight Line Equations for the Individual Plots.

 

2121:. a
1 0.962
2 1.286
3 1.313
9 1.285
5 1.119
6 1.103
7 1.212
8 1.955
9 1.936

10 1.372

 

Illllilih

Oh

....v

...-5.!

- \n ‘5‘

.‘Ah

:3: x 131:: a. y

.3h

9.. N «N ~Su~

99

.mp0Hm moon“ moofioow no: on woowymw>oo

 

 

 

newcomer one mowmso>m xooow open one .muoao Edmozwno osu mom mooao> Home“ open one moons :
. SS .Bssaxsoeoo norms m
.owmooomws ahnmeou nommm ooowmoz may you unomon oosowdnomno .mnma anchuosm gonad m
.omma .xom use seems a
m.oa m.HH m.oa m.n o.HH o.ca .>.o
o.o I o.m I H.o I m.: I ~.o I w.m I .a.m
r.sm+ o.om+ ~.mm+ o.mm+ o.om+ o.om + .oaa
floss some sees more memo floss mm .ama
Ammo hoes Ameo Amos Aces aces mm .saa
m me He mm am an ac mm ca
m me so am mm mm mm . mm m
OH me so me am so so mw w
m mm mm Ho Ho mm mm as a
s om Hm 3m mm om Hm mm o
m a: m: mm mm m: a: mm m
mg on :m we Hm mm mm cm s
o co co co co cm so am m
ma on :m we Hm mm mm om N
H m: m: we as m: w: m: H
owe uon

owns»— — cognac—9% _oonw:9§m=‘.ownaoefinom powwoofihwm 02 e Hon cone—3E8...

 

 

m nuseaxoo>ou- m coupons , a xom use spasm were

 

.mowoohm noose one mesh pom oomoao>oa wo>eoo thm
oasesroo spasm mo mosao> xenon roam ream use so nonaeedsoo .>H oases

 

(II

'II
' .

 

50

Site index values were determined for each plot by reading
them directly from the harmonized and polymorphic site index
curves deve10ped for this study (Figures 5 and 6). The site
indexes fbr each plot were also determined from other published
and unpublished site index curves by Gevorkiantz, Shetron, and
van Eek, in Table IV. When all 10 plate are combined it can be
seen that the various site index curves yield similar average
Site index values for red pine on Rubicon sand. On a plot to
plot basis however, the site index values are not always similar
and this is reflected in the standard deviations and the ranges
among the site index curves for individual plots (last colum,
Table IV). The range in the predicted values agree to within one
site class at age 50 years but, disagree before and after age 50
with the amount of disagreement increasing as the plots become
younger or older than age 50. Hewever, when van Eck's curves
(deve10ped fOr different soils) are omitted the ranges with the
other curves are within 1 3 feet, or one site class. Spurr (1955)
and Alban (1972) have also found that red pine has a relatively'
uniform growth pattern over a wide range of sandy soils.

Because red pine is not growing on the same soil in each of
the aforementioned site index studies the question arises as to the
uniformity of the growth patterns. van Eck has studied red pine
growth on several soils ranging in texture from loams to sands.
His published growth curve represents the average of these different
growth patterns. Figure 7 compares van Eck's average curve to other
studies which concentrated on red pine growth on the sandier 30113,

After age 35, van Eck's curve bends more sharply than the others.

51

 

 

 

:3_,
D
l\
C{_.
O
to
‘3. ..
O
In
A
on C3__
3 s
\“."/
.- o
' r-
6: 8
-
'5! c».
o _
N
/ This Study
van Eck
o. _ Mosinee ...—.—
3 Lake States ----
T—
o J I I I I I I.
TL] 10.0 20.0' 30.0 90.0 50.0 60,

AGE (years)

Figure 7. Comparison of the Mean Growth Curves from This
and Other Studies.

52

van Eck has shown that this sharp bend is characteristic of the
loamy sites and is distinctly different from the sandy sites.

It can be seen then that any site index curve deve10ped for red
pine must describe the different growth patterns red pine exhibits
on the different soils or groups of similar soils.

Polymorphic growth curves were developed by van Eck to more
adequately describe the growth patterns of red pine on the different
soils. .A comparison of the polymorphic curves of van Eck's study
and this study, in Figure 8, Show again that the growth rates
of red pine differ widely between the two studies. There is a
marked difference especially in the early ages. In his discus-
sion, van Eek states that the growth pattern of red pine on Rubicon

sand is different than his published curves.
Another method which can be used to evaluate site quality is

total volume growth. The comparison of volume growth fer this
study has 3 shortcomings: (l) the stand density, or basal area,
varies widely from plot to plot, (2) the tree ages on each plot
were not identical, and (3) some plots had been thinned and
others had not. Little can be done about the first problem.
Problems 2 and 3 however, were overcome by constructing volume
growth curves fbr each tree. In this way, all trees could be
compared at the same age, either before they were thinned or,
after the thinning age had been attained, the prediction of
volume growth for the unthinned plots was based on the volume
growth pattern of the thinned plots. For this study, age 50
years was used as the reference age because it was the age at
which the tree heights were compared. Caution should be used

when interpreting the volume data because the wide range in basal

53

 

 

 

O _. SI 70
s /
/
O _ [.jl SI 50
s .z/
.x/
o/ I
c: I SI 50
SI’ / .4”
... /
/”‘\ ‘1’
'3 °. I . SI 90
o S” /
"' /
V
I- / I
O
I "'
o 5’» .
III
I ‘2...
8 This Study ..--..
van Eek -.-.-.
‘2
o P-
...
q—
c; I I I I I I I
0] 10.0 20.0 30.0 90.0 50.0 60.0

AGE (years)

Figure 8. Comparison of the Polymorphic Site Index Curves
develOped for This and van Eck's Study.

50

area renders equal comparisons impossible and secondly, individual
tree volumes are greatly affected by stand density.

volume growth should be compared on a per stand, or per acre
basis and not on a per tree basis, because as Spurr (1952) states:

...the factors which affect growth per tree are not

necessarily the same factors that are most important

in affecting growth per acre. The diameter and other

dimensions of the individual tree are greatly influenced

by the competition of its neighbors. The effects of

root and crown competition frequently outweigh other

factors related to growth of a single tree. On the

other hand, growth per acre is very largely a function

of site quality. On a given site, a dense stand of

many small trees will frequently have about the same

volume as a fairly Open stand of the same age consist-

ing of a few large trees.
Since basal area per acre cannot be determined for previous ages,
by actual measurement, a basal area of 120 ftz/acre has been
assumed for the reference age of 50 years. This assumption is
considered reasonable because according to Buckman (1962) it
represents a basal area of medium intensity management, and
varies little from the actual basal area of plot 3 which is
50 years old. The data are summarized Table V. To derive the
data presented in this table the following assumptions were
made and are considered reasonable for a stand 50 years old:
basal area of 120 ft2 per acre; 200 trees per acre} diameter
outside bark of 10.2 inches; fbrm class of 90 (this is the
average form class of the trees in plot 3, which were 50 years
old); form factor .uz (suggested by Gerorkiantz and Olsen for
trees older than 30 years old).

Table V shows the volume per acre determined by the formulas
or tables deve10ped by 3 additional studies and the volume for

this study as determined from the graph in Figure 9. Using the

.~sma ..num .mxm .pom museum mxmg
.mmma .xoonucwz anumunom
.33 ..fim .mxm ..Sm man—3m 83

HIV!“

.mo:aw> cwumfiwumm +
.m no woman can wawnuo on» mm mamas mommy : zazo co comma mw muoam omega pom mwwno>w 0:9 as
.mamo> om gasp mama none was unmum .muoam moon“ pom manuawm>m no: was sump “cowowmmsm s

 

 

55

 

 

 

 

A.ma a.ma m.HH m.HH .>.u
mm: H mmm H mmm h :om H .a.m
a.smmm o.o:a~ o.osa~ m.mmm~ mwmnasc
samm oso~ seam maam so an «sea
mamm swam osmm Hamm mm mm m
mama comm comm ammm as mN +m
amen ooow comm Hamm mm as a
mmsm osm~ ommm smm~ Hm em a
oHsN o-~ comm ssmw a: mm ««m
I I I I mm c~ a:
mmmm omwm omdm Naam on on m
I I I I mm cu «m
swam scam o~m~ mHNN w: m: H
m :msxosm numuw saga ~ assume a numaoamm .mam .mmq. scam
Nahum fiXhO>uw

 

.mw< mo mass» cm as scam sum mo masao> poem usage amquHumm mo nomanuaeoo »> magma

56

60.0 80.0 70.0
’1 GT

40.0

Mean Curve .... ....

HEIGHT (feet)
20.0 30.0

10.0
1

\

{In

0.1

 

 

10.0 20.0 30.0 40.0 50.0 60.0

AGE (years)

Figure 9. Harmonized volume Growth Curves Using the Mean Growth
Curve of All Trees Combined as the Guiding Curve.

VSI Vblume Site Index

57

method prOposed by Gevorkiantz and Olsen, and that prOposed fer
Canadian forests resulted in the lowest variation between plots.
Both of these methods are based on the taper or form of the
average individual tree. When the method proposed by Buckman or
this study were used to determine volume per acre much more
variation between plots resulted. Buckman's method utilized the
relationship of site index and basal area to volume growth, whereas
this study utilized the relationships of volume per tree with age.
The average volume per acre of the 0 different methods
ranges from 2529.9 to 2960.0 ft3/acre. These volume averages
are relatively similar indicating, like site index, that red.
pine growth is uniform on Rubicon sand. The volume per acre
growth between plots however, can vary considerably depending on
which method is used. The Canadian and, Gevorkiantz and Olsen's
methods assume a uniform taper of all trees, whether they occur
on "good" or "poor" sites, and the variability in volumes per
acre is only about 11% which means that volume and height growth
are equally sensitive to site quality. By removing the affect
of uniform taper, data in Buckman's study and this study show
the variation is almost double the variation of site index
between plots, and indicates that volume growth is more sensitive
to site quality than site index. The two latter studies are
probably closer to actual field conditions because a uniform
taper would most likely not exist. However, this was not verified.
In a similar study of red pine, in 1969, Hannah found volume
growth to be a more sensitive indicator of site quality than site
index. waever, whether he used site index or volume growth he
rated the site quality of Rubicon sand for red pine as medium.

58

Soil has been shown to be one of the most important factors
affecting site quality (Carmean, 1975). To determine its effect
on site quality selected soil pr0perties were compared to site
index using stepwise regression analyses. Site index was used
although it is probably less sensitive to site quality than
volume per acre. It was chosen to be the dependent variable
because tree height was measured precisely and accurately in the
field whereas, the calculation of volume growth was based on
many assumptions which, although they were considered reasonable,
may or may not have been true for Rubicon sand. Therefore, fer

this study, site index is preferred as the dependent variable.

Soil Data

Variation in site productivity due to soil differences is
well documented in the literature (Coile, 1952; Mader, 1968;
Hannah and Zahner, 1970; Carmean, 1975; Page, 1976). Their work
has shown that site index can be correlated with one and some-
times more than one soil prOperty. These soil properties fall
into three broad catagories viz., soil physical, soil chemical,
and soil topographic features. In this study 31 variables were
studied (they are listed in Appendix III). The following soil
features will be discussed in detail:

a) Soil texture

b) Depth to water table

c) Gravel content

d) Landforms

e) Soil nutrients

59

a) Soil Texture

The presence of a fine sand layer (band of fine sand or
finer) or slight increases in the amount of sand, barely detect-
able in the field with ones fingers, usually resulted in a better
site.

While lodking for suitable areas for plot establishment one
area.was found that had a loamy sand surface texture. This is a
finer texture than that of Rubicon, which is sand, and as a
result has a higher water holding capacity. This difference
was reflected in the increased conpetition. Because of the
increased competition the red pine growth.was inhibited. Red
pine probably did not reflect the higher potential because the
preper weeding procedures were not undertaken. Thus, the presence
of a fine sand layer does not insure a highly productive site.

The affect of finer soil layers can also be seen in the site
indexes of red pine growing on Graycalm soils, in Table IV. These
soils are similar to Rubicon in all major respects except for a
fine sand band and thin, more clayey, horizons between 90 and 60
inches from the surface. The average red pine site index on the

two Graycalm sands is 71.9 whereas, on Rubicon it is 55 to 59.

b) Depth to Water Table
The water table depth could be accurately determined on only

2 plots. At plot #2, age 20 years, the ground water was found at
a.depth of 99 inches (in July) with no apparent drainage mottles
above it. At plot #9, age 39 years, the ground water was en-
countered at 106" (in August) with drainage mottles beginning

at 75". The site index for the two sites are 55 and 60 respectively.

60

For the remaining 8 plbts the upper surface of the ground water
was not encountered, even to a depth of 10 feet in all plots.
Because the water table was encountered at only the two plots,
little can be inferred about its relationship to site quality
from this study.

van Eek (1958) found that red pine is greatly affected by
the soil water regime. He found that the site index of red pine
at the foot of a slope was generally much higher than at the mid
or upper slope positions. Red pine found on other sandy soils,
e.g. Croswell, which have higher water tables than Rubicon,
also had higher site index values. However, he was not able to
discern the exact affect of the water table because he found
that other soil properties varied concommdtantly. For example,
finer soil textures at lower slape positions and decreased soil
depth on the steep positions had an unknown affect on the site
index.

In a more extensive study, the Perest Service has found that
the water table, even to a depth of 17 feet, has an affect on
tree growth (Personal commication with Richard Watson, U.S.I-‘.S.,
Forest 8011 Specialist).

c) Gravel Content

Thin gravel layers as such, were not found on any plot.
Gravel pockets however, occurred frequently and.were usually
associated with root proliferation throughout the gravel pocket.
In the field it appeared that these gravel pockets enhanced the
site quality but, in the final analysis of the data, the total

gravel content was found to be inversely related to site index.

61

If roots proliferated in the gravel pockets why does the total
gravel content negatively affect site index? van Eek (1958)

also encountered a curious relationship between gravel content
and red pine site index. He states that, "apparently the effect
of gravel is of a complex nature: partly one of benefit to trees
on coarse textured soils if it was associated with a textural

B horizon, partly one of harm to growth if occurring in quantity
and larger sizes near the surface, not associated with a finer

textured horizon.”

d) Landforms

'Most study plots were located on sandy outwash plains.
Three plots, #9, #8, and #9, were located on border areas between
a sandy outwash plain and a moraine, a lake bed plain, or a
drainage course spillway, respectively. The site indexes of
these 3 plots are 55, 6M, and 63. The average site index of these
plots is slightly higher than for plots which occurred only on
sandy outwash plains. Although the soils in each plot are
classified as Rubicon sand,in the border areas where the plots were
located, the effects of the soils in the adjacent areas can be
seen in the slightly higher site indexes. These differences
were attributed to the presence of a band or layer, from less
than % inch thick to as much as l or 1% ft. thick, of finer
texture ranging from fine sand to clay loam, found in the adjacent
area. It is possible that the tree roots were able to reach
such areas in at least 2 of 3 plots and obtain more water and

nutrients, but, this was not confirmed.

 

- luau-LIILHI-fi. .-

62

One of the original objectives of this study was to compare
Rubicon sand in terms of red pine site quality on different
landforms. The official description of the Rubicon series states
that it occurs on ”tills, outwash and lake plains, moraines,
and to a lesser extent on old beach ridges and sand dunes....”
The author believed that if Rubicon actually occurred on such a
wide variety of landforms it could not be used as a reliable
measure of site quality. This objective had to be abandoned
however, because Rubicon which was supporting red pine could
only be found on sandy outwash plains. Only on one occasion
was Rubicon found on a landform other than an outwash plain. It
was found on a moraine but it was supporting a stand of mixed
hardwoods and not red pine. When the different landforms were
checked it was found that other soil series usually predominated.
These other soils possessed either a higher degree of podzolization
(Kalkaska), finer textural subsoil bands (Graycalm), thick finer
textural subsoil bands (Montcalm), or finer sand (Rousseau), etc..
Rubicon sand could only be fbund on two different types of land-
forms: moraines and outwash plains. Based on this study alone
it does not mean that Rubicon occurs on only these landforms but
it does indicate that care should be taken when describing or

interpreting the range in landforms on which it does occur.

e) Soil Nutrients

The fertility, or nutrient status, of the soil is difficult
to estimate because, where trees are concerned, the rooting zone
is only vaguely understood. Knowing the volume of soil occupied

by the roots during the different stages of tree growth, viz.,

 

-wuxnmm _

63

establishment, rapid growth, and mature stages would facilitate
the estimation of available soil nutrients. The picture is
further complicated by the fact that forest trees can recycle
nutrients (Curlin, 1968) . To estimate the fertility of Rubicon
sand, the soil profile samples from the study plots were analyzed
for the following nutrients by the Michigan State University Soil
Testing Laboratory: phosphorus, potassium, calcium, magnesium,
percent carbon, and soil reaction (pH) . The results are summarized
in Table VI. The quantity of available nutrients was in the fol-
lowing order: .QB’R’JS’M- The ratio of the percent carbon in
the A, B, and C horizons was approximately 6:2:1, respectively,
and the pH of the soil increased from the A to the C horizon.
Stepwise regression analysis was employed to find the
meaningful correlations between site index and the numerous soil
and non-soil site factors (See Appendix III for the soil and site
parameters used in this analysis). Site index instead of tree
height growth was used as the dependent variable because its R2
value (coefficient of nultiple determination) is a better estimate
of the site quality variation as explained by the independent
variables (Carmean, 1975). Using tree height as the dependent
variable is less desirable because its close association with
age masks the affect of the other site variables (Carmean, 1975).
For exanple, Table I has shown that for this study 92.0% of the
variation in height is associated with age alone. This high
degree of association means that if all of the other site factors
are combined they could only be associated with 8.0% of the
variation in height growth. This masking affect of age can be

6n

 

 

 

  

 

w.m 5:4 n.mm m.m: :ImH m.~H n.mm c.oa H.~N +oouo: o
s.m mm. m.mm a.ss o.o~ a.ss m.mob a.m~ H.m: osIm m
H.m mm.~ a.:a n.nw ~.m. ~.m~ A.anm 5.3: :.aa NI: d
ed Esau ” . h 8 a m 323
unoonom momma .soxo Humoamm show no season acumen Hwom
.o2¢

 

.muoam ca Add mo sunbeam any one sosau> omega .u:oawao=
.uasm savanna mo munuanusz wanuwausoxm saw manaaausa mo snueasm .H> magma

65

overcome if tree heights are compared at only one age. Since site
index, by definition, is tree height at a given age its use as

the dependent variable is justified. In this analysis, the site
index values were taken from the harmonized curves deve10ped for
this study because the harmonized curves exhibited less variation
between plots than did the polymorphic curves.

The results of the stepwise regression analyses show that

m!) .id U.‘ V]

to explain up to 71% (r2 = .70977) of the site index variation,

5 variables were necessary (See Table VII). These variables,

 

in order of their degree of correlation are:

1) pH of the B22 horizon

2) Thickness of the A.horizon in inches

3) Basal area

0) Percent gravel in the profile

5) The average percent fine sand in the A plus B horizons
The equation for this relationship is:

Y I 12.36 + 5.52(X1) + .259(X2) +.OIIS(X3) - 360(4) + .30(X5)
where,

Y - Site index

Xl- pH of the 822 horizon

Xz- Thickness of the A horizon

X3- Basal area

X48 Percent gravel in the profile

x5- Average percent sand of the A plus B horizons
This equation is presented only to show the relative rank of the
variables found to be most significant in this study. It must

be kept in mind that this study consisted of only 10 plots and

66

memos.

mammm.

mommm.

Hmmmi.
Hmmms.

 

mmxvoom. + 352;. I

Rooms. + Amxvmmm. + 3535 + 3.3

ass a? . I

333m; Q53... + $53.2 + 33.?

$539+ 33%. + 253.3 + :36?

33?. + 5.53:3 + 25.97

90392 + En?
IIIflmISBs m

muonwaoz m + 4 sou mo
0 mm perm ocwm unoonmm owwao>< m

I Hm Ho>mao unschym :
I 5 32 H38 m
I 5 52.8: 4 25 no $ng N
I S 55.3: «.3 at mo .3 H

3935 «3385 .mqum

 

 

:moum

nose um Gomono moanmwnmai "moshing

cowemohwum omwsnoum mo massesm .HH> manna

67

to state that this equation represents the true relationship of
the dependent and independent variables would be misleading. It
can be used however as a guide to the site factors which affect
tree growth and ultimately the site quality.

As an example of how this might work in the field, plots 1
and 9, with site indexes of 08 and 63, respectively, have been frtҤ
compared. Using the five factors found by this study to be

related to site index the data is as follows:

 

Plot #1 Plot #9 '
pH of the 822 = 5.0 6.0 g N
Thickness of the A.horizon = 8 8
Basal area = 70 90
Percent gravel in the profile = 13 5
Average percent fine sand (A+B)/2 . 16 39

Next, substituting these values into the formula,
Yplot 1 I 12.36 + 5.52(5.0) + .259(8) + .095(70) - .36(13)
+ .30(16)
= 95.3
Yplot 9 a 12.36 + S.52(6.0) + .259(8) + .045(90) - .36(S)

+ .30(39)
- 61.5

By using only these soil factors plot 9 would be shown to be a
better site than plot 1. All of these soil factors, with practice,
can be determined in the field.

Soil classifiers, responsible for defining the range of soil
characteristics for a given soil series should be aware of what

factors are important for tree growth and should describe the

68

soils in such a way that these factors are taken into account.

For example, the range in pH for Rubicon might be redefined or
further restricted, or the range of the amount of gravel that

can occur in a Rubicon profile, which is not Specifically stated

in the offical Rubicon series description, could also be defined.
Because small differences in the amount of gravel may have an
affect on site quality the range of gravel accepted could be
restricted to, say, less than or equal to 5%. The exact percentage
would have to be determined from a larger number of samples. Soils

which would then contain more than 5% gravel could be set

 

aside as a separate phase of Rubicon or another series. Or if,
after examining a large number of samples, the establishing of a
Rubicon phase based on the amount of gravel, could not be justified,
other factors such as pH of the 822 horizon or thickness of the

A horizon should be considered.

It is equally important for foresters to recognize the soil
and non-soil factors which affect the site quality. In ameliorat-
ing poor sites it is important to understand why the site is poor.
Is it a result of edaphic or non-edaphic site factors? It has
been argued by many that basal area has relatively little affect
on site index except in extreme cases (Carmean, 1975). Data
from this study supports this argument. In the previous equation
basal area (X3), was one of the factors found to affect site
index even though its affect was small: + .OQS(X3).

Being able to recognize some of the more influential and

subtle characteristics of the soil such as the pH, percent fine

69

sand in the subsurface horizons, and thickness of the surface
horizons will help the forester, as well as the soil scientist,

identify the effective site factors.

 

CONCLUSIONS

Conclusions from this study are:

l) Harmonized site index curves can be used to represent
the growth of red pine on Rubicon sand, when stem
analysis is used to determine the growth pattern. The
growth pattern is adequately represented by the mean
curve and little, if any, advantage is gained, in terms
of management, by developing polymorphic curves.

2) Site index for red pine can be determined from a number
of published site index curves. This study shows that
the difference in site index values between these site
index curves for any given plot, can be considerable.
However, each set of curves yields a similar average
site index for red pine, i.e. 55 to 59, for the plots
studied.

3) Red pine growth curves are not uniform over a wide range
of soil profile textures. More site index curves such
as those of this and van Eck's study should be developed
for groups of similar soils.

0) When volume per acre is used as the indicator of Rubicon's
productive capacity for red pine the results are similar
to those when site index is used. Different methods

show similar average volume per acre growth of red pine

70

 

5)

5)

71

on Rubicon sand (2520.9 to 2950.0 ft3/acre), yet, the
variation between plots can be considerable (11.3% to 19.9%) .
Site index and volume per acre are about equally sensitive

to Rubicon site quality for red pine if a uniform taper

is assumed for red pine. However, if the affect of uniform
taper is removed the sensitivity of volume per acre almost l_flҤ
doubles (i.e. from 11.3%.to 19.9%9.

Of the properties of Rubicon sand and other factors of the

site, those most correlated with variations in red pine

site index are: pH of the B22 horizon, thickness of the

 

A horizon, basal area, percent gravel in the profile, and

average percent fine sand in the A.plus B horizons.

LIST 01’ REFERENCES

 

LIST OF REFERENCES

Alban, David H.

1972. An improved growth intercept method for estimating site Frrfia
index of red pine. NOrth-Cent. For. Exp. Sta., St. Paul
Minn., (USDA For. Serv. Res. Paper NC-80).

Anonymous.
1938. water tables and survival. Lake States Forest Exp. Sta., ‘
Tech. Note 139. r

 

Barrett, J.P. and L. J. Goldsmith.
1973. Predicting growth of eastern white pine. New Hampshire
Exp. Stn., Durham, N.H., Sta. Bull. 999.

Buckman, R.E.
1962. Growth and yield of red pine in Minnesota. Lake States
For. Exp. Sta., PS, USDA. Tech Bull. 1272.

Buol, S.W., F.D. Hole, and R.J. McCracken.
1973. Soil genesis and classification. The Iowa State university
Press., Ames Iowa., pp. 360.

Carmean, Willard H. and Alexander Vasilevsky.

1971. Site-index comparisons for tree species in northern Minnesota.
N. Cent. For. Exp. Sta., 8p., illus. (USDA, For. Serv. Res.
Paper NC-6S).

Coile, T.S.
1938. Forest classification: classification of forest types with
special reference to ground vegetation. J. For. 36:1062-1066.

1952. Soil and growth of forests. Adv. in Agron. 9:329-398.

Cooley, John H.

1970. Thinning and fertilizing red pine to increase growth and
cone production. N. Cent. Forest Exp. Sta., St. Paul, Minn.
8 p., illus. (USDA.Forest Serv. Res. Pap. NC-92).

Curlin, J.W.

1968. Nutrient cycling as a factor in site productivity and forest
fertilization. 125 Proc. of the Third N. Am. For. Soils
Conf., pp. 313-325. .

73

79

Day, N.H., C.P. Bey, and V.J. Rudolph.
1960. Site index for planted red pine by the 5-year growth inter-
cept method. J. For. 58:198-202.

DeBell, D.S., A.P. Brunette, and D.L. Schweitzer.
1977. Expectations from intensive culture on industrial forest
lands. J. For. 75:10-13.

Della-Bianca, Lino, and David F. Olson, Jr.

1959. Site index comparisons for several tree species in the
Virginia-Carolina piedmont. Southeastern For. Exp. Sta., ”‘2“
Asheville, N. Carolina. (USDA, For. Serv., Sta. Pap. 109).

Doolittle, Warren T.
1958. Site index comparisons for several forest species in the
southern Applachians. SSSAP 25:(5) 955-958.

Ferree, M.J., T.D. Shearer and E.L. Stone, Jr.
1958. A method of evaluating site quality in young red pine ‘
plantations. J. For. 56:328-331. “’E

 

Freese, Frank.
1969. Linear regression methods for forest research. Forest
Products Lab., For. Serv., USPS Res. Pap. EFL-17.

Gevorkiantz, S.R., and Harold F. Scholz.
1999. Determining site quality in understocked oak forests.
J. For. 92:808-811.

Gevorkiantz, S.R., and Olsen, L.P.
1955. Composite volume tables for timber and their application
in the Lake States. USDA.Tech. Bull. 1109.

Gevorkiantz , S .R.
1957. Site index curves for red pine in the Lake States. Lake
States For. Exp. Sta., USDA, For. Serv., Tech, the 989.

Hannah, Peter R.
1969. Stemwood production related to soils in Michigan red pine
plantations. For. Sci. 15:320-326.

Hannah, Peter R. and Robert Zahner.
1970. Nonpedogenetic texture bands in outwash sands of Michigan:
their origin, and influence on tree growth. SSSAP 39:139-136.

Hanson, Thomas J.

1975. A computer technique for projecting yields of even-aged
red pine in the Lake States. Coll. of Nat. Res., Univ. of
Wis.-Stevens Point. 59981.

Heiberg, Svend 0., and Donald P. White.
1956. Site evaluation concept. J. For. 59:7-10.

75

Heinselman, M.L., and Z.A. Zasada.
1955. A review of the literature relating to quaking aspen sites.
Lake States FOr. Exp. Sta., Sta. Pap. 32.

Hills, G.A.
1952. The classification and evaluation of site for forestry.
Ont. Dept. of Lands and For., Res. Rep. 29.

1959. Field methods for investigating site. Ont. Dept. of Lands
and For., Site Res. Man. 9 (revised 1957).

"""'"‘1960. Re_'g_iena1 site research. For. Chron. 35Iu01-u23.

Husch, B.
1963. Forest measurements and statistics. Ronald Press. 979 pp.

,Miller and Beers.
1972. Forest mensuration. The Ronald Press.

Lemmon, Paul E.
1958. Soil interpretations for woodland conservation. In: Proc.
of the First North Am. For. Soils Conf. pp. 153-1582

Locke, Stanley 8. .
1991. The use of soil site factors in predicting timber yields.
SSSAP 6:399-902.

Lutz H.J.
1958: Geology and soil in relation to forest vegetation. I_n,: Proc.
of the First N. Am. For. Soils Conf., pp. 75-85.

Mader, D.L. and D.F. Owen.
1961. Relationships between soil prOperties and red pine growth in
Massachusetts. SSSAP 25:62-65.

Mader, Donald L.
1963. volume growth measurement - an analysis of function and
characteristics in site evaluation. J. For. 61:193-198.

1968. Soil; a guide to forest growth patterns and maximum produc-
tivity. Mass. Ag. Exp. Sta., Coll. of Ag., Univ. of Mass.,
Amherst. , Exp. Sta. Bull. 570.

Martin, Helen M.

1955. Map of the surface formations of the southern peninsula
of Michigan, Geological Survey Division, Michigan Dept.
of Conservation. Pub. No 99.

 

76

Mason, Richard R., and Timothy C. Tigner.

1972. Forest-site relationships within an outbreak of lodgepole
needle miner in central Oregon. USDA For. Serv. Res. Pap.
PNW-196, 18 p., illus. Pacific Northwest For. and Range
Experiment Station, Pertland, Oregon.

Minckler, Leon S.
1993. Effect of rainfall and site factors on the growth and
survival of young forest plantations. J. For. 91:829-833.

National Oceanic and Atmospheric Admin., Environmental Data __~ H
Service, National Climatic Center, Asheville, N. C. (No date). 1

Page G.
1976: Quantitative evaluation of site potential for spruce and
fir in Newfoundland. For. Sci. 22:131-193. ,

Ralston, Charles W.

1958. Some applications of soil evaluation in southern forest £~W
management. lg; Proc. of the First N. Am. For. Soils — .m-
Conf. pp.159-166.

 

Retzer, J.L.

1958. Soil - a factor affecting the distribution and growth of
native and exotic forest vegetation. In: Proc.of the
First N. Am. For. Soils Conf. pp. 110-115.

Rowe, J.S.
1953. Forest sites - a discussion. For. Chron. 29:278-289.

Shetron, Stephen G.

1969. Variation of site productivity among soil taxonomic units
in northern lower Michigan, Ph.D. Thesis, University of
Michigan, Ann Arbor.

1972. A study concerning the soil growth relationships of native
jack pine and red pine plantations on Mosinee Paper Co.
lands. unpublished report, Ford Forestry Center, Mich.
Tech. Univ.

“1979. Distribution of free iron and organic carbon as related to
available water in some forested sandy soils. SSSAP
38:359-362.

Soil Survey staff.
1975. Soil taxonomy. Soil Conservation Service, USDA, Ag. Hand-
book 936. pp. 759.

Spurr, S.H.
1952. Forestry inventory. Ronald Press, New York.

77

Spurr, S.H. ‘
1955. Soils in relation to site index curves. SAP Proc. 1955.,

pp. 80-85.

 

1969. Forest ecology. The Ronald Press.

Steinbrenner, E.C.
1973. Mapping fbrest soils on Weyerhaeuser lands in the Pacific
Northwest. In: Proc. of the Fourth N. Am. For. Soils

Conf. pp. 5131-525. [W

Thomas, B.R., and E.R. Burroughs, Jr.
1973. Significance of soil classification in forest land manage-
ment. lg; Proc. of the Fourth N. Am. For. Soils Conf.

van Eck, Willem A.

1958. Site and root studies of red pine (Pinus resinosa Ait.)
plantations in lower Michigan. Ph.D. Thesis, Michigan
State University. --""'

 

, and E.P. Whiteside.
1963. Site evaluation studies in red pine plantations in Michigan.
SSSAP 27:709-719.

Wakeley, Philip C., and Jose Marrero.
1958. Five year intercept as site index in southern pine plantations.
J. For. 56:332-336.

Wambach, R.F., and A.L. Lundgren.

1965. The importance of site quality in red pine. Papers of the
Michigan Acad. of Science, Arts and Letters, Vol. (50)
pp. 67-7“.

Weber, H.L., R. Hall, N.R. Benson, and G.V. Winter.
1966. Soil survey of Grand Traverse county, Michigan. Soil Cons.
Serv., USDA., in c00p. with Mich. Agr. Expt. Stat.

White, Donald P.
1958. Available water: the key to forest site evaluation. lg;
Proc. of the First N. Am. For. Soils Conf.

, and Robert S. wood.
1958. Growth variations in a red pine plantation influenced by a
deep-lying fine soil layer. SSSAP 22:179-177.

Wiggans, 0.0.
1969. Depletion of subsoil moisture by apple trees and other
woody species. Nebr. Agr. Exp. Sta., Res. Bull. 216.

Wilde, S.A. and D.T. Pronin.
1999. Growth of trembling aspen in relation to ground water and
soil organic matter. SSSAP 19:395-397.

 

 

APPENDICES

APPENDIX I

- t—e-f
refill: u 3..

 

I...

7E3

RUBICON SERIES

The Rubicon series consists of deep excessively drained soils forced in
sandy deposits on till plains, lake plains, outwash plains snd moraines. These
soils have rapid peraeability. Slopes range from 0 to 90 percent. Mean annual
precipitation is about 30 inches, and mean annual tenperature is about 93° F.

Igggnggig_g1ggg: Sandy, nixed, frigid Entic Haplorthods.

Ignigg : Rubicon sand - on a 3 percent convex south facing slope in
a red pine plantation. (Colors are for moist soil unless otherwise stated.)

A1--0 to 1 inch; black (101R 2/1) sand, flocked with light brownish gray
(10TH 6/2); weak fine granular structure; very friable; coaeon roots; very
strongly acid; abrupt snooth boundary. (1/2 to 3 inches thick)

A2--1 to 6 inches; light brownish gray (101R 6/2) sand; very weak nediue
granular structure; very friable; cannon roots; very strongly acid; clear secoth
boundary. (2 to 7 inches thick)

021ir--6 to 10 inches; dark brown (7.5!R 9/9) sand; weak aediua granular
structure; very friable; sany roots; eedlua acid; clear wavy boundary. (9 to 12
inches thick)

0221r--10 to 18 inches; dark yellowish brown (101R 9/9) sand; weak coarse
granular structure; very friable; cos-on roots; aediua acid; clear irregular
boundary. (0 to 20 inches thick)

B3--18 to 36 inches; yellowish brown (1OYR 5/6) sand; very weak coarse
subangular blocky structure; very friable; aedius acid; chunks of ortstein occur
at depths of 18 to 29 inches and represent about 15 percent of the surface area
of the horizon exposed; chunks are 9 to 6 inches in disaster; colors are
yellowish brown (1OIR 5/6) representing 60 percent of the sass and dark reddish
brown (SIR 3/9) and pale brown (1013 6/3) representing the resalning colors,
sand; passive; few roots; weakly to strongly ceeented; eedlue acid; clear
irregular boundary. (9 to 20 inches thick)

c1--36 to 60 inches; light yellowish brown (1OYR 6/9) sand with acne coarse
sand in upper portion; single grained; loose; slightly acid.

Iypg_nggtggn: Cheboygan County, Michigan; 0.5 ailes west of Highway M-31
on Hackleburg Road, then 200 feet north in the 891/9, 8'1/9, sec. 5. T. 35 9.,

I. 1 H.

ngngg_1n_gnggggtggigtigg: The thickness of the solue ranges free 20 to 50
inches. The reaction of the solun ranges frca eedlus to very strongly acid.
Coarse fragments range to as such as 5 percent throu hout the solum. Mean
annual soil teeperature is estiaated to range free 930 to 97° 1".

The A1 and Ap horizons have bus of 101R or 7.SYR, value of 2 through 9 and
chroea of 1 or 2. The A2 horizon has hue of 10!!! or 7.5111, value of 5 through 7
and chroea of 1 or 2. It has weak granular or weak subangular blocky structure
or is single grained. The A horizon is send or loaay sand.

The Hair horizon has hue of 10th, 7.5!R or SYR, value of 9 or S and chrome
«of 3 through 6. The 83 horizon has hue of 101R, 7.513 or SIR, value of 5 or 6
and chroaa of 9 through 6. The aeount of ortstein occurring in the Bar and 83
horizons range free 0 to 20 percent. The structure of the 02 and 83 horizons
range from weak granular to weak subangular blocky or it ls single grained.

The c horizon has hue of 101a or 7.5rn, value of 6 or 1 and chroma or 3 o
9. It to medium or coarse sand. The reaction ranges from slightly to nudinn

79

“Udldnfl JERI€3--?

acid.

Competing_§gnign: These are the Crivitz, Croswell, Deerton, Duel,
Graycalm, Karlin, Kiva, Rousseau and Vilas series in the same family and the
similar Deer Park, Grayling, Kalkaska and.wallace series. Crivitz soils have
finer textured sola. Croswell soils have mottling at depths between 20 and 90
inches. Deerton soils are underlain by sandstone bedrock. Duel soils are
underlain by limestone bedrock within the control section. Graycalm soils have ~.m_
Bt horizons. Karlin soils have loamy fine sand or sandy loss in the 10 to 90
inch control section. Kiva soils have stratified coarse sand and gravel at
depths ranging from 10 to 29 inches. Rousseau soils developed in fine sands.
Vilas soils are developed in medium and coarse sands containing a higher
proportion of dar colored minerals, including slates, schists, iron bearing
rocks and red sand tone. Deer Park soils are spodic intergrades. Grayling
soils lack spodic horizons. Kalkaska soils have a Rh horizon more than 3 inches
thick. Wallace soils have a continuous ortstein.

 

naggggpnig_§gtting: Rubicon soils are on till, outwash and lake plains and
moraines and to a less extent on old beach ridges and sand dunes along the Great “
takes. Slopes range~froa 0 to 90 percent. The mean annual precipitation is 21 E_,_;,
to 33 inches, and annual temperature is about 90° to 95° I.

gga‘n.nn;g‘111_‘..391.33g_§911g: Croswell, AuOres and Roscommon soils form
a common drainage sequence with Rubicon. lalkaska, Drayling and Montcalm soils
are common well drained associates. -

: lkcessively drained. Surface runoff is slow.
Permeability is rapid.

flg._gng_lgggtgtign: The greater proportion of this soil is forested,
including tree plantations. Some areas are idle cropland or in permanent
pasture. Only a very small proporlion is used for small grains and hay crops.
The native vegetation and present uatural vegetation is dominantly red pine and
aspen with some halite and Jack pine . Ground cover consists of blueberries,
wintergreen, sweet fern and reindeor moss.

. MW: lor'hern half of lower Michigan and mper
Michigan. The series is of large vxtent.

W: Ontonaso' County. Riemann: 1922.
nglgnxg: The Rubicon series las formerly classified as Podzols.

Rational Cooperative Soil Survey

03".

uuumnumcamnmmaue

80

GMYCALH 8"!”

The Graycalm series is a sandy, mixed, frigid Entic Haplorthods. Typically these soils have a
very dark grayish brown send Al horizon, dark brown and strong brown sand IZir horizons,
yellowish brown sand 33 horizons, light yellowish brown sand A2 horizons and yellowish brown
and with bands of brown and reddish brown lomay sand It horizons. ,

£121ch Pagan: Graycalm sand on a slope of 1 percent on an outwash plain in n forested
area. (Colors are for moist soil unless otherwise stated.)

Al--0 to 3 inches; very dark grayish brown (10!! 3/2) sand; moderate medium granular
structure; very friable; many fine roots; very strongly acid; clear wavy boundary. (2 to 5
inches thick) '

lZlir-3 to 6 inches; dark brown to brown (7.5!! 9/9) sand; weak fine granular structure;

very friable; co-on fine roots; stronglyeacid; clear irregular boundary. (3 to 12 inches
thick) ' ‘

IZZir—6 to 13 inches; strong brown (7.5Yl 5/6) sand; weak fine granular structure; very
frifile; few fine roots; medium acid; gradual wavy bomdary. (0 to 12 inches thick)

iii-~13 to 22 inches; yellowish 'brown (10!! 5/6) sand; sinle grained; loose; fmv fine
roots; slightly acid; gradual wavy boundary. (0 to 10 inches thick)

A2~22 to 35 inches; light yellowish brown (10!! 6/9) send; single grained; loose; very
few fine roots; slightly acid; abrupt broken boundary. (10 to 25 indies thick)

AID-35 to 60 inches; light yellowish brown (101! 6/9) sand (A2); single grained; loose;
lanes and bands of brown (7.5?! 5/9) and reddish brown (STR 5/9) loany sand (It); weak
medit- subangular blocky structure; friable; bands are ll9 to 2 inches-in thickness with a
total accuulation of 5 inches; 5 percent by veins of pebbles; slightly acid.

W: Clare Colmty, Ilichigem; 2310 feet west and 700 feet north of the southeast
corner of Sec. 6, T.“ 20 l., k. 9 H.

gas in Charactergtig: The thickness of the sol:- ranges from 90 to greater than 60 incl-u.
' Dqths to the It horizon ranges from 35 to 96 inches. The pebble content throughout the pedon

ranges fra 0 to 5 percent by volt-a. The null. rangms from very strongly acid to slightly '
acid. The mean annual soil temperature is estimated to range from 99' to 97". The Al horizon
has hue of 10!! or 7.513, value of 3 or 2 and chrome of l or 2. In cultivated areas the &
hgrizon has hue of 10!! or 7.5Tl, value of 3 or 9 and chrome of 2 or 3. Some pedons have an
A2 horizon, l to 9 inches thick. It has hum of 101! or 7.5Yk, value of 6 or 7 and chroma of 1
through 3. The A horizons are send or lo-y sand. The Dir horizon has hue of 7.5!! or lDYk,
value of 9 or 5, and chrome of 9 through 6. It is send or loamy sand. The 33 horizon has hue
of 7.5Yl or 1023, value of 5 or 6 and chrome of 9 to 6. It is send or buy sand. The _A_2
horizon has hum of lOTll, value of 6 and chroma of 2 to 9. The It horizon is in bands l/l6 to
2 inches thick. The total accuulation within a depth of 60 inches is less tha 6 inches. The
ht horizon has hue of lOYl, 7.513 or 5T3, value of 9 or 5 and chrome 9 thrth 6. It is In
sand or light sandy loam. Some pedons have a g horizon with hue of 10", value of 6 or 7 and
chrome of 3 and range from slightly acid to mildly alkaline.

may Segig ad Their Differentiae: These are the Crivitz, Croswell, Deerton, Dual, kiva,

Pomfret, louseau, Rubicon, Seney and Tiles series in the same family. and the unless, Lealaau
ad liontcalm series. Crivitz, Croswell, Deerton, Duel, liva, Pomfret, Rousseau, Rubicon, Bessy
and Vilas soils lack It horizons. Chelsea soils lack spodic horizons and are masic.’ Laelanau
and Montcalm soils have argillic horizons.

Setting: Graycalm soils are on till plains, moraines and outwash plains of Visoonsinmn age.
Slope gradients range from 0 to 35 percent. The climate is continental, with a mean annual
precipitation of about 30 inches. The mean annual teqerature is about 93'!. and the nan
sl-ar teqerature is about 65”. .

 

 

enmi-

81

GRAYCALH SERI PIS-"- 2

firlnrlnal Associated Soils: These are the nearby Rubicon, Seney, and Crnyling soils on the till

plains, moraines, and outwash plains, and with Montcalm soils on till plains and moraines.

"Ella-1111.:'Qlllloflniziilili'slyl_l.l[3.2 Sum-what excessively drained. Runoff is slow or very slat on the
Ill-.lrlv Ivvvl nluln-z: mul mu-dlum um llw slut-[11W (lllcfl. l'rnncnlvillly ls rapid.

llm- .mll \‘l-w'ldllinl: A lmgc- purl Is: In (urn-:l land. The Inna-ll vvizu‘lul‘lnil (‘nnzllnha ('Illi‘lly of
«ML and hlrkury will: mmw wlnllr plnv In llw :muihurn part, and jnrli NIH! and scrub nnk in tho
nullhrrn part «I the nrvn. A low whllv plnc are in some nrcns. A small part is cropped to
small grains, corn or hay.

Distribution and Extent: This series occurs in the central and northern part of Lower flichigan
and in the eastern half of the Upper Peninsula. This series is of moderate extent.

 

Sprics Pronosed: Gladwin County, Michigan; 1966.

Remarks: The Graycalm soils were formerly classified as weakly developed Podzol.

National Cooperative Soil Survey
0. S. A.

W‘A‘CS'tM‘fl-N. "I. ”7.

IS ....‘m- (h.

 

APPENDIX II

 

i- -.vv°-- .

82

ea no «53352 on no.5 Sofie? 888 in no cog—no Ba 3.
x... he b.3325? on foo ”.339, eeoooe 3: no .8332 2:. e

 

 

 

 

 

 

 

 

seasom. sham. sswcmm. Hmam. ouna
esmmmm. mama. Hmmm. Ammm. ca
owmm. n:nm. uhhm. huum. m
mmmm. waww. swam. mama. m
namm. :Hmm. muma. mmmo. n
mmom. m3cm. monm. mama. c
ammo. :mna. samba. Nana. m
mwom. swam. cosm. wwmm. :
ssmmmm. mswm. seasaa. mnmm. m
nmom. mnmm. ssmhhm. mem. N
one“. one“. eoewm. mono. H

NxU + X: + m u w xn + n n w Nxo + x: + w I > xa + m I.» uoam

flung. pmeoeo zoom 3 3%

cocoon 98335 do no.“ contiguous mo ouconoflmoou

d mHnma
HH XHazmmmd

83

 

semmcm.
senamm.
ssaowm.
ssnmwm.
ssoamm.
esaanm.
sswmwm.
ssaamm.
scammm.
sesamm.
esmomm.

ea he ascends? on nmxxt Sofia? 383 one no 833% one 2.
ca u... ufioflnnuo on amxbv 3&2? 383 an mo 533%. 29 e

 

mama. bemoho. :~:c. cAIH
Hmmm. ssoowm. ammo. OH
uhsm. eewmmm. mmms. m
omcm. ekmomm. 33mm. m
om3m. esmmwm. mmsm. n
mNNm. ssmwmm. mmmn. o
NHNm. ssmmnm. Nwmw. m
Nnmm. esm33m. :Nm@. 3
mama. esmmmm. mmmm. m
mmHm. sswomm. mwom. N
mums. esmmmm. swam. A

 

Nxxao +

 

 

 

>.moa xwwn + n I w mod NX\wo + X\wn + w u >.woa Xumn + n=a > moa noun
9 mum scum nmmwtummnn

89

r- “a
T

Hot: can one. he poeoflwcwem i.

 

.uouoonon mu
wwwozuomhn ecu .mnnu muwnmonwaw>nso mo nonwoo unwowmwcwwm aagwwz n msozw awnmnowumaon unu momma :uon
:H .ocwa unmwwnum w ma own one pgwwo: mo mwswcowumaon can “can mos swoonuonzn Hocawwno may

eHe\H eeoh.omaa ~x\wo + x\wn + e w won

x\wn + n u > wed

 

 

 

 

 

 

3e) 33.? ~xo + E + e u s
3 + e ... >
Eoooonm mo moonwoa nuoam gas now :owunnuu Gauchos oaam> Ml cowuonum
A“ muoan.aam pom camps o moonwa

auwnnonfiq pom munch can no abusesm

m manna
HH Xanmmm<

85

 

 

Hoeon.xn ago he unmanagewem ee

 

~X\Ho + X\Hn + m I > mo.“
x\wne+ m I >.woa

NKO+£+MI>

xn + e u s

 

mHo\h~ ssaqco
o~©\m.n :9... mm
cHo\NN 94.6.... N:
o~o\ma ssmm.~o
Eoomonm mo mom on 9.59, m

on“ no numoam one 3025 consumes nogummwa 05 guess

   

3.0.3 .3315..ch

 
 

eonmmmmw _..::

.eonuoeeu one: naeeo>o use me been no“: ecowueeum hone Heeeu>nocH

U manna
HH Kan—Ema

.oonmwnmcrou mo nonhfloé mo g

86

Fr“ 5

53303 mo 9.39on we

Ho>oa x: any so peeoemwewwm ee

 

 

 

 

oem\wa eemn.os ee~\e eene.o~ ~x\wo + x\wo + e u >_wos
num\on eemm.H~ . me~\ee eeeo.m x\wo + e u w won
oem\ha eeee.m somxe eemm.m Nxo + xn + e u w
nem\ca eemH.HH mo~\e ee~n.ns xa + e u n
.mlo .mwlafl Mm undue, m nag
n «o «m a: .N «H moose on 4wfllm .m ouonm

. macaw Hon—.3595 o>wuoommom 5.9.3 on 60.39.50 5:3 macaw 9:395 .3393 on“ now go
now: on“ one 30.3 mason”. noumnm 93 no.“ 0% can: any mo saganmcroo mo 323.an mo hams-56

a mafia.
HH Kan—g

HEIGHT (feet)

87

APPENDIX II

Table E. Comparison of Individual Plot Growth Curves to
the Mean Growth Curve of All Plots Combined

30.0 60.0 50-0 80.0

 

/ Overall Mean curve (all plots) .... -_
Mean curve (this plot) —.—-_
Estimated from overall mean curve

10.0

 

 

00'
F

L l_ l J 1 1
of “no 20.0 so.o 40.0 so.o no.0

AGE (years)

PLOT #1

 

F,

APPENDIX 11
Table P. (continued)

HEIGHT (feet)
20.0 80.0 90.0 .0 00 .0 70.0

10.0

 

88

 

 

Overall Mean curve (all plots) __ ....
Mean curve (this plot) ..--—
Estimated from overall mean curve

 

l l l l I l

o.' 10.0 20.0 so.o 40.0 com com

AGE (years)

PLOT #2

7 't

89

APPENDIX II
Table 8 (continued)

”-0
l

00-0
1

00-0
I

 

90.0
I

HEIGHT (feet)

 

Overall Mean curve (all plots) __ ......
Mean curve (this plot) _..._
Estimated from overall mean curve

10.0
I

 

0'1
I

I l. 1 _li l .1 l
o.' 10.0 so.o so.o 40.0 60.0 so.o

AGE (years)

 

PLOT #3

90

APPENDIX II
Table B (continued)

70 .0

 

40-0

HEIGHT (fast)

20 .0
1

 

Overall Mean curve (all plots) _ _.
Mean curve (this plot) ...-.—
Bstimated from overall man curve

10.0

 

J 1 L 1 1 J
o.' 10.0 20.0 30.0 40.0 30.0 no.0

AGE (years)

APPENDIX II
Table I: (continued)

HEIGHT (feet)
20.0 30.0 40.0 80.0 80.0 70.0
1 T

10.0

 

0.1
L.—

91

    

/

I Bverall Mean curve (all plots) .... _
Mean curve (this plot) m-
/ Estimated from overall mean curve

I J l l l 1
10.0 20.0 ”.0 40.0 80.0 00.0

AGE (years)

PLOT#S

 

APPENDIX II

92

Table E (continued)

HEIGHT (feet)

20.0 30.0 40.0 ”-0 00.0 70.0

10.0

 

 

 

I

I Overall Mean curve (all plots) __ _.
/ wan me (this Plat) -'-_
Estimated from overall mean curve

1 l 1 l L l

 

10.0 20.0 ”.0 40.0 30.0 00.0

AGE (years)

PIAOT#6

APPENDIX II
Table E (continued)

HEIGHT (feet)
20.0 00.0 40.0 50.0 00.0 00.0

10.0

 

00'
-+—

 

/

1

10.0

93

 

Overall Mean curve (all plots)
Mean curve (this plot)
Estimated from overall mean curve

1 l 1 l L
20.0 ”.0 40.0 00.0 00.0

AGE (years)

PLO‘1'#7

9”

APPENDH II
Table E (continued)

70.0

 

60.0

HEIGHT (feet)

 

/ Overall Mean curve (all plots) _ ....
Mean curve (this plot) ...-.--
Estimated from overall mean curve

10.0

 

0"
F.

I I I I I I
0.' mm 20.0 00.0 40.0 no.0 00.0

AGE (years)

PLOT“

95

APPENDIX II
Table E (continued)

70-0

. .‘llr‘.

 

HEIGHT (feet)
30.0 40.0 60.0

 

I... / Overall Mean curve (all plots) _ __

Mean curve (this plot) -.....
Estimated from overall mean curve

10.0

 

0.J

I I L I I I
0.l 10.0 20.0 00.0 40.0 00.0 00.0

AGE (years)

 

PLOT #9

96

APPENDIX II
Table E (continued)

70.0
*1

 

40-0

HEIGHT (feet)

 

20.0
9T

.. [If Overall Mean curve (all plots) ____ ___
Mean curve (this plot) _,._
Estimated from overall mean curve ___—____.

10.0

 

0.1

I I I I I I I
0..1 mm 20.0 00.0 40.0 30.0 00.0

AGE (years)

PLOT #10

APPENDIX II I

 

Ian-
HowmummcwmI—I
O 0 0 0

12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
29.
25.
26.
27.

28.
29.

30.
31.

97

APPENDIX III

Variables Used in Stepwise Regression Analyses

Thickness of the A horizon
Thickness of the B horizon
Thickness of the A + B horizon

pH of the A horizon

pH of the B21 horizon
pH of the B22 horizon
Depth to textural bands

Percent gravel in the profile

Percent slope

Aspect of slope (NE, N, S, SE, SW)

Thinned or not thinned

 

Landform (Sandy outwash plain - SOP; SOP bordering on Moraine, g _.., .
Lake Bed, or Drainage Course Spillway)

Basal area

Depth to maximm percent fine sand
Percent fine sand in the A horizon
Percent fine sand in the B horizon
Percent fine sand in the (A + B)/2 horizon (Average)

Average pounds per acre
(A 4- B)/2 horizons

of

Average pounds per acre of

(A + B + C)/3 horizons
Average pounds per acre
(A + B)/2 horizons

Average pounds per acre
(A + 'B + C)/3 horizons
Average pounds per acre
(A + B)/2 horizons

Average pounds per acre
(A + B + C)/3 horizons
Average pounds per acre
(A + B)/2 horizons

Average pounds per acre
(A + B + C)/3 horizons

of
of
of
of
of
of

available phosphorus in the
available phosphorus in the
exchangeable potassium in the
exchangeable potassium in the
exchangeable calcium in the
exchangeable calcium in the
exchangeable magnesium in the
exchangeable magnesium in the

Average percent exchangeable potassium in the (A + B)/2

horizons

Average percent exchangeable potassium in the (A + B + (2)/3

horizons

Average percent exchangeable calcium in the (A + B)/2 horizons
Average percent exchangeable calcium in the (A + B + C)/3

horizons

Average percent exchangeable magnesium in the (A + B)/2

horizons

Average percent exchangeable magnesium in the (A + B + C)/3

horizons

APPENDIX IV

 

98

 

 

.mxm 53 88.5203 mm .93 a} ”as: 300.com 230 .253 pm 3008 8.30% no 00:03;

3.3 g» V $9.08. >30 .30 a} E m; an. #0 H\m.~ 5:: .3302. 23 2 303:3 H2055
:moa H4 mm gem H< "Ewmnmyao any new: umowconpw .uwm uzogwnoasp msonwucoomwu ocean Umaoaoo
“Ewan— "..omIo: pm 533 A 55:3 .mofiwcou 5 .Xm V 5.33.5 25%?! .33 who mm nomad EH
muoom mammuds um RH V mmmfincoum ":31. um RH v mmmcwcoum “noaoo nonmembumm .5350: H4

 

 

 

 

 

 

 

g;

is:

i

.3... .8. m2 082

I 9583: mm” :83

$3-3 82.584 +33 .8... .3 Susan
seizing G052“. :03 ......5 wfldfi Manama 3....

 

 

5.8 £0.38 035m £35...

Guam 3.8.... 2!!

10550080950— 030 a 0:3 some 433353 .m.w .80 33......
03a 8 ”28.... a znom.~ 5:0 .2» .m as ex... “a. 388 a H3 0% 59c 8.83
I33: 035 .335 €me .065... 3.4.0.8

 

 

 
   

 

  

 

 

 

H ..z .24 GQRG a... .awéxwuuuflaJN 00m 32. £33 mo 3m :2 mm (Hal
a: 98m 2 ..m- 8023. .8...

”BE comm as 80.5 .302

>H Kan—Ea

 

 

 

99

 

umscaucou at poem

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.0 m.o m.o on m o\o awed +mm we
ma. H.c .mae 3w o.o on w u\w saga wmuaaa mmuam Ho
o\a mama
om. o.o _mae 3w o.a 0H m e\w_m»oa rcauumupo\z_am-m mm
o\u gouéommmv mucus
mm. ed 30 o.m 3 m m\m ES; 3.64.33
om. o.m so o.m on m :\5 gym oaum naawm
1;. ms: 3m o.m 0H m ~\m gm mIm Nd
an.fi o.m as m.: on m ~\m m>ca ~\u «was AMIHV H4
mso
S a.) 8.8. 8.0: to
“u :m n. at. :9. ages axle ii
.238 ...... » ......
X nag W 8.5128 8.8

 

 

 

 

 

 

 

 

 

 

 

100

L .1], 5i,

...} ME: 40 new: 0.5 as.
xowzu EEm v momma ads :33 .....c .GHE :3 .CH3 ..mm um 25 Jan EH .5000 moans m

 

 

 

 

 

 

 

 

£38!!!
also-......

8.8.»

:m: ~3me I wmmwcon. .....ie 88. mm .082

Romuma I neon—muse 23.3: x~%&oa
mmu+=m~m=m~on§Tm 32.8..— ..m... 5...... 83......
..1..8:8_ .0052“. .39: 62.8 mmflaoh may? 5....

 

 

:33 Songs >UEmm 202832..
cram 5.8.... it...

 

 
 

 

 

 

 

 

 

 

ésg=ofl no . :88 8v .! .z
...on B 2:34 :23 3:2. cm 3. on. a 3.. .0 ...on 52m 8:83

“55:01.0: anew .ud.u.n.uu .uun._._a ”Gnu... 33.8

H ..z 83— 33> 08—.8 .mnuxwtofia 4.2.3.2. .3600. .0 www.mmqmm .(hal

N... 98.. .2. .... zooms”. a»...

ZO_.—L.¢§ 4.0a

v0.5.3.3". >H “992qu

10].

 

353:8 mt SE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i. o. am a.“ 3 o m 3% ”E: .5..— Eu
3. cat um 30 a; 3 o m ...}..m 6:: 91.2 S
2.. mg. am .5 a: 3 o m .3 Es: ~m..m~ mm
34 NA. am .5 m6 3 o m 3} 6:: 3-2 m3
3. fin om so m.m 3 o m m} 6:: 2-3 gamma
mo.~ mi Eu 3w m.m OH 0 m :\3 63.5 Edam hwamm
£4 mg mo mm 9m 3 o m ~\m 5:: m4 2
m .. H
NH.m m.w* ME mm m}... OH G gamma H\m EH N\.: gm: HO d
flu
u m a... :3 to 23 to can
. .38
.338 6.: 928 23...... ...o...
mm Aw m i 8.90 L

 

102

 

.3an o:

€37.33 2 2.} ”E: .30 6:3 .53 2 3.} ”E: .md .ufin ..on 2 353m
8an £3385 “:8me mm .33 2 xi. 5:: .30 .ufin :8 “E 2.} 6:: JG
.ufin =8 2 23,. 53 6d .ufin ....m 2 Ed ...} En: ...3 93mm 33 ESE

Ll

 

 

 

 

 

 

 

 

£853..

8.15

5.3.989: 32 882

x . 2338. «$4. use.»

Rm Gm...“— mmma 822.8»— ...cmNH 883.8 3:26.»
..3.33_ 22. 82.8 wfldg maufiqflfi.

 

53m 5838 £8.35...
65m 5.82. 2.2...

Hmucwcwucoo E nowumuzmam 2H3 Hi .88 8...! .2
“on 3 2 mm a 35m 55 .358 B 3 ELflmeaflm: a 2 ma non 28m .883

 

 

 

  

 

 

 

693.3." a: 035 3 Em co 6 8:91.818
A .0283.— fixfih :8 8 fiumxwwflfifl as. .233 .Hm .omm mo xszimdmgltz
m: .85 .3. ..m I 2833. an...»

29.—Lian- 45w

Umafiucou >H Xanuhmd

103

 

umszwuaoo m# uoflm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘H3. ~.c om m.a ca o m m\s m>oa +=mm_ on
'1'

m:. H.@ om 3@ m.m ea c m e\a «yea wmumw do
on. ~.o om 3m m.m ea c m o\m_¢>m.n m~-- nHNNm
~o.a o.o on so o.m ea c m :x: mwm.n NH-“ awamm
oa.H m.m am so o.m ad a m :\m «yea 5.: m<

Hm.a 5.: so_ 3a m.: ea c .m ~\m.¢>o~ :.a Na

ms.“ 0.: as; an m.: ea c m H\m.~ «yea H\n_m>ad Hue Ha

cud do

u =m ...-8: 3.2523 .8: an ..8

_x nag mm raaxéfixz. as.“ .aea» Lana .2ro 43:

nuns-no V ;,

 

10H»

.m c5 98 .umfim ..Nd E
3h ”5: 5.8 in: 2 mm} Em: .mfl .mwfiuoe Ea .3} E3 .8.“ .283; ”E :5
“E mm\m ES ..mnnw “Hm mo wmm .xonmmm .sz x ..m .xonmmm mmflwnop 53930 .9.“ H ..2

£368.32

 

 

 

 

 

 

 

bloc-El

8.15

wnwz3mwam Rm :93 mon 633: So: 32 «983
$505 omnwa .5322 wmmna 8.5
58m §Tm 35:2...— +=md 53.5 ‘ 5.23
=3...3:8_ yawn“. :95 8.5.5 wmwflon Spa .13.

 

mcwmpozéfimam £92550 32.—mm Ema awaken conga-5323..
unmm 5.82. 22...

Hmucmcfinoo 32:3 mmfifimv 53358 SE xomn a 8m .83 3.3.2
“3m 3 3:3 :23 ES) 3m oo .3 8:98 on m E 3),. ow max 3 w mm you 52m 8:83

 

 

 

 

 
 

 

 

 

605.33 m: 3:qu 3 Hum uwaE c cot-$330
H .281 onH> 38—3 .3 33an sawmififimm 03 mo $32.32.? {L5
i 93%.... ..z 2823. 25.8

ZO.E.¢OQUO Java

333:8 B finzmmfi

 

omaawuaoo :# «can

 

 

 

105

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No. ~.o om m.o ea c w o\~ m>oa +m: mo

so. o.o om 3w o.o_ ea c m m\c m>cH m:-~H do
mm. ~.m co 3w o.e oH . o m cxm awm.h ha-m nfiwmm
om.d m.m _== 3o o.m ea c m :\:_m>m.h mum “flamm

mm. m.m .s: mm o.m ea c m m\m m>ca mum ~¢

om.4 o.m hm mm o.m ea c m ~\s «wad ~\m myoa m-o H4

cud Ho

mg.“ may} ...... 3...: 3.3 s in...

 

 

106

 

.wufin mm 0: 22-2:
2 HEB mm 9:8 =2: 3 ma 2 3.} EB msosfipaooflu 2 “fins 5:: mommunofi
3:3 mo mofiufifi .3} 5m; .Em 3 mi 3.879. 59a mfiufin 1.55:. maosfiucoofln

 

 

 

 

 

 

 

 

 

 

 

 

as;

w: in!

93% 8.8a

magnum mums: muwxoom 5 :wnndooo muoon 05mm 52 £53.. 89¢ mlz 882
23.3: XNIH 83v.

Rm :34: 822.81 .33 :93 ago: 53.8 cousin
__1233_ SEE Has 33.8 5&2 Samoagfl 3....

55.8 nwmzuso 3.5m Eat-35....

 

cmmw 3.82. 2.2...
HmHEwcHusou 8258— :ouumpnma wann tum 320 B. a... .z

 

 

 

.83 3 25:} H pm :23 3 E Shm N ow - 0% mo .358 mm 59¢ .333

 

H .02 gmfl

 

 

 

m# 98m .02 0E

205.8080 ...00

3:55:00 >H Una—g

107

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

umsufiucou mu uoam
q... 4:”. am .2. 3 o m oxmé ES ...: +~m mu
.umw .395
9.. 9m am 5 ad 3 o m ab 5:: WAN 8
mm. «5 cm 3 ca 3 o m a} 53 Tum n33
£3 9.”. .5 an m.m 3 o m 3} Es; TN n33
ANA. flm me an oz". 3 o m 3} 52 NA 3
85 NJ. me mm A; 3 o m A}. ME: .333 To 2
. a #3
o-H do
u .3 m a... .8: b5 23 23.8... to £38 ..chn:
A3 3.8... .88
x R III-Io 3.8 {

 

 

 

 

 

 

 

 

 

 

108

 

.ucwmnw wcnmn +=-H pd “mmsoum 553 .083 ..«E .238; :NNH OH
.3: 2 3 co: “2 3.8 ..mm 2 “mm. 65; :3 2 mm} ME: 6% .3 cfin =3 2
2.) ES; ...: x :3 5338 =- ..E Eoflmnmmflona poop fit. .oommm 3325 amino

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3...;

Din-IE8

8.15

w 13.8..— m 683

xm 55 mama +§~ ‘ 23......— fi .8.»
x31.“ :34: 88.88— :wma 55 39.1 .838 Santa
aging-L . 3....

, 5.3m :m 23.91....

we? 8 m mmmrnom new a 3.8!. :2...

Hmufifiucoo .828 ..wfi Exams 3 3m 2 3\m.~ . 53353 «:3 Extra 83:...
.ucmficnufluouwsoamgnmfia guard: . ow : 8:83

 

uofipowmw. Sufi dumfié€mfis .35» 8.338

~ 9.88— 2\c\m ..8 .3 9.3 £3 3mm :32. .5 8m mo .mmzdmiz .(huu

 

 

 

on 88.2 3: 83mm. .33

203L502 4.00

"wanna—n00 >H XHDE

109

 

 

cmacwucou ow yoga

 

 

 

 

 

 

 

 

uh. H.o om o.o 0H o m :\c m>oa +=w~ Ho
om. ~.m am as o.o ad a m mxm «sod .w~-~a m~m
mm.H o.m am so m.m ea c m e\m_m»m.n ‘hauo na-m,
mm.a :.m go on o.m ea c m :\m_m>m =u-m QHHNm
3.4 m.m E E o.m 3 o m ~\m a» m; 3.4 .2
ma.:u
mmwm m.m as 3m m.: ea c m H\n m>ca cauc H4
“magma a paum a~nau
and Ho
0 mm to 8.8. 39!
can 5 :8
m J 25. 3 to:
‘ L b

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.10

.5 mo 33 .m Ea mm .umflm $332: 2 "mm 63qu
13-3: 2 23,. 53 am .2 65.3 ..mo E 3} EEK .md .2 65a 3o :8 2

of;

 

 

 

 

 

 

 

 

 

0.38.5!
8.25
55138.. w 683
+=._._N Rm 563. mws 2330! $4.. COO-m
......NIm Km 08...:20— +mNH Gar—fl whg ~30! :0 .3803
__1..8=om_ 35.an #33 £2.20 5:8 maulmuagma 5....
53m :835 Exam ......g
\fimflf its-E 2.2!.
328 3.58—

 

    

3208. it. .2
”BE 8 no “2 m ow 55 .2 3}. H 2 cm :23 2 3\m H m cm Hfl a N - m you 58m 8.53

costs a: 035 3 Em 852. can

so
A 6283— om\w\m zoo—ma. mung mmflm .33— .mem. .mm owm mo mEm ..sz3m w“
.E
3" 93m 62 a:

 

 
 

200 Hg."— 83 =3
20:280qu 45¢

vuflfiucoo >H finzummd

111

 

Umflfiwunou m# padm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3“. o.o om m.o OH a m :\a myoa +=:Al o
as. m.m am so c.e ea c m c\w «>QH. 1:~-mL-m
no.H m.m am so m.m .oa o m :\5 m>m.~ maum pHH~a
mm.H m.m as am o.m oH o m m\: «woa muo ac
o-H Ho
3
u xgmgssfig __ $5: 3%..
um awn mm nilllao .830 , M .
i/T

 

112

 

 

.Sm B flu 35a Edge: 2
we} 53 dd .832... Binge: 2 3}. Es: 6d .3 B 3 23 ..mm 2 no.» - o2
Hmooamsu mo mm "995 25.25 sums mm: wand u~\m gm ...: x :m mw and um wawaou :wwuwuno

 

 

 

 

 

 

 

 

all;
if

8.15

.28... 8.... dz 682

l can... «3 .8.»

Xm 52$ 2:: ......safi gm: 52: mung 38.5 8:28
318...; 3:33 :2» .2520 Jammie.“ 3.2.

 

:23 can 93 so $2253 538 £358 556 283......
“and 3.82. .33..
Hancofipaow 32.8. . 53353 2E com goat-2,...

 

 

23 3 :8 m 55 E S} m a 32 NM a Hm you 58m 8:83

 

 

 

3:23 a: as
a 288% 2);} 38—8 2853 .95.. 95% Ex z-s - 8m 32 3m 3%. mm

3.8.3.22: 95035 25.8

 

 

 

 

 

 

 

203L530 4.0a

vagina—co >H “DH—Ema

113

 

 

8:528 3 SE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

u H6 on ed 3 m m\n SH T: 3
mm. #6 co 36 oé 3 m :\m EH #5133
m... 9... om so m.m 3 m m} 53. T _mem
morn o.m oo 30 o.m OH w 3}. 53.5 T7133
"2 3
E25 m.m me am oi. 3 m w} ”E: -53."an H 9%
~\m EH «XHE .3ka

tOIN HO

0 mm S to .3: a!» .38. be 83: to :8
x as Mia-... ea...» 5.89 to:

. R lull-Io 8.8

“TI

 

11.4

 

.u 8:: mnowfim m 5:3 33$.” 5953 @835: $5 EEK .35 633.8. ..mn 2

g;

 

 

 

 

 

 

 

 

 

 

 

 

aisle);

:\3 gm :23

e873 2:3 ..m Ba 53.... :3. 3m .83
:SNIOH aim—“OH fiwuwuho 23:03 mam 82¢.
mam .xonmmm 32.38— ..ooa .835 cousin
:3... 82-1 32.35 Jame.” 33....

GHmHm :wmzuso 33mm uiog

ll! 65m .282. ...-....

flucmfiunou 32.8— , 53353 ES uwm .83 33:...

 

SE B :2. t :23 2.3m macaw com 2 cm 8% w omamém pan 52m 5.83

 

 

-505

H .ozasm— 25$ .8 o E
E

88—
$ 8.: .3. ..z

ZDUHmDM 25 so»

ZOEEOONO 4.8

3:528 3 5%qu

115

 

 

3:528 2, .33

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.. as am .3. 3 o m {a 5:: +1. _rdu
fl. :6 om so. od 3 o m a} .5: 3.8 mm
mm. Nd am so as 3 o m a} ”E: 84$ 36%
mm. ad 8 8 as 3 o m m} 66H 3-2 3&3
9.4 m.m 8 so Wm B o m 1.5.63. 24 p22
i. m; .5. 3m 9m 3 o m N}. 66H TA 2
ii .34 ms 3. 9m 3 o m ~\m «>3 L5 2

aoua Ho
mkmzfiifiiisfig ...... in...
. 8.8.1.8 3.8 w
J sull

 

 

 

 

 

 

 

 

 

116

 

 

.30 .2 .253 +82 2 3 fl? 55 28. gum 5:3 68mm 353 63 .3 .85;
com “2 gown“ :H ... a» .3 zany“; 9.".ka Hwoohmnu 393:0 noaoo o: ...: x a: Jaw—.5»
8223 Exams :5 2 E}. ”E: ...fi can. ...S ...8 pm 35a 3 3859583 2

 

 

 

 

 

 

 

 

ii

5.82....

in

.38.... m 6&2

..oNTNN swam.“ .85 23...... x: .3...
..Ilwlaludujo m a ......sfl LEW 5:3 E
=9..- 83% 00—5..an .333 ’58 >HHH£ i

 

323 mmaadmm - umgwwwafinnéfid £838 35m 3.5.1....

 

a .8803: «:2...

 

Hmucmfificou 82.3 . nouumunmam 05m cmx .88 8.68.:
90.3 on now m :93 fish—w 2:: 3 we a}. .n m mm 33m a omamnmm pod Eonm 8:83

 

 

 

uofinflmwm Bu. 5 gave 6%? .uccmm £43m
«mm . .

a . , \. go.» ham « «2.3a. own 0 mm . 32
. £88— . o \o~\m 1876.8 co 3 5 33¢ S m x
3* 8.3.4... ..z 2828. 25.8

 

ZOE-1030 4.0a

doflflwflfloo >H Xanumn—d

 

 

 

cue—3:8 3n uoE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

my. :6. am od 3 o m :3 6:: +2.. 8

mm. N5 om m.m 3 o m a} ”E: 213 m2
7 3. ad 8 m.m 3 o m :\m «SK :1: 5.0.2

n

8.: m.m 8 ad 3 o m i: 63K 34.. p28

34 m4. 9.. m.m 3 o m 3} «>3 m..m 3

$4 m... m... m.m 3 o m m\m ”E: To 2

o-H Ho

0 19 mm... 603 «.3 8 s 68

x :3 m .8... 25» can: .38 .5:

R ...-.128 38
\l

 

 

118

 

. 655m .8 mo 8.3 .32 2 353.... 98 coma 2
.3332..." m.8 283m .53 2 3.}. .53 .3“. 22.83 .3 .35.“. ..mfl-..om 2 £3 53
£339: :2 2 $33.3.m 3.5% :23 .083 :oflmgwmfloam poop :85: 2 335
...H ..uwzma HmouHmao ..mHQm H4 mm: xowm Hmnflxwg pcwumnmm o: 25 museum €53 .oommm
23:3 mg ”E: 9:8 ...} NE: 6d .23 ism :35 EN £83... .2 .mufin =mm-..om 2

 

 

 

 

 

 

 

 

8.3.1.8!!!
DSQIEIC

5&5“

.5wa mo Rm .23.. .8: fig!

:24: 533.5 22...... gm 83.

Km Ems... mme 32.8“: :mma :93 pmnwmopw ...-3.5 cos-in
=5...8=8_ uwcwmpu 3w: onus-5 92.30.“ Spam 5....

>m3aawmm I wwg sflmwmsfimnn 29533.:

 

95m 3.82. is...

Hmucwfivcou 895.3 :owumuqmam mam cum 18.08.6212

uowa com 2 :93 ...—um u :23 3 mm: on wE S\m.m m o mmamumm w um mm #0. E m 5:83
1 cozunoamm= owucm .Ewwpm .wawE £35m co: .8

 

    

 

 

 

 

H .283 3):} aid. 95.5 .626 .302 .52 .8 0% mo $32 mm 25
.DU UhanwB
a; HQE 6: oE éuwgw 25:8

ZOCE¢930485

“5:55:00 >H “3an02

 

 

119

 

 

uosawpsoo Hax “can

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_ww. n.m m.u ea c m o\@ «yea +mm Au
- m.m am 3% m.o ea c m o\m awed _mmumN -m
or. ~.m am so o.o ca 0 m :\3 awed _m~sma awamm
ou.a H.m as am m.m ea c m H\:.m>oH iwauoa ~<
Hm.a m.m «a ma o.m ea c m m\m_m>ca oauo m4
cum Ho
u mm m a. 8.. 83...an 23 3o: ..8
x as 658.8... .25» 23.8» 5.80 to:
Mm 3:31:30 .330 {H
xtxul.

 

 

 

 

 

 

 

 

120

 

.~\~ ES. 1 in 66 ...m x =8 2:98p £9.38 38.53 L can H8220 ..m 2
3328.5 .m 3 .~ 5 3258 J62“ a.) 3 is .u~m-mfl .Hmufl £23 =39~m 2

 

 

 

 

 

 

 

 

 

 

 

 

at};

is:

cow-ohm

55.... .8: mm 683

IIIEE 3 8:5 23%: Km . H .83
Km :93 mwma .8585,— +w~a 53.6 8:265
......obzom— \ Umfiwmpu .353 00-595 HEW—”Hon Hubw 3:3.

.353."wa .. mmhsou wwmcwmpo xi-gf

unmm 5.8-... :2!

Hmucmfipcou 3:52 , pod” B m 5N ”8335““ SE c353 332....

 

55 .2 mac? com m :23 £83 83 B we Ska mm 33.2 w 3 dog 5.83

Uosynoammm owycm .UfiwHMM .UwXHE .Nunmm co: :0

H ...z 83.— Rhmé 3% 96,6 >295 :53 .23.... .3 8m mo £32.32“ .wu
.3 23%

Ease 25:8

 

 

 

Nat scam 6: 2:
205.580 ...3

Umsflwucoo >H XHQZMEAE

121

 

umsswuaou NH: uoam

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

- 3.0 am c.o ea c m :\c m>oH +flmauagi do
Hm. m.m am so o.o ea c m o\m m>oa a:-m~ mm
o“. N.m am so a.w ea c m o\m gym.“ wwuad nw-m
m~.H h.m om so w.m ea c m :\s mwm. twaum uHHNm
om. m.m so 3m o.m ea c m ~\m mwm.n mum ~<
on.a ~.m 20 mm m.: ea c m m\m.m>ca mno g4

and do

0 mm mm to .8: .03 to 23 3°! :8

x nag m .23- 38. .928 238» 8.8 .anoofl .5:

J/arlu.

 

APPENDIX V

 

APPENDIX V

122

 

 

:FfiRLTP”—"

 

 

 

 

 

 

 

 

 

SUMMARY 0P LABORATORY ANALYSIS DATA
Plot Nutrients A B C A B C
Aye lbs/acre Aye percent
Exch. Bases
P 9.5 71.3 11.0 — - -
1 K 13.0 16.0 13.0 9.3 26.3 28.6
Ca 228.5 119.0 - 81.9 73.2 —
Mg 210.0 12.7 10.0 15 9 55.5 71.9
P 7.0 26.3 15.0 - -
2 K 25.5 9.7 - 3 3 '2.1 -
Ca 900.0 171.3 286.0 78. 72.1 91.1
Mg 92.0 39.0 15.5 17 7 25 9 8.9
P 29.7 69.0 27.0 — — -
K 29.7 32.0 19.0 3 9 9.9 6.9
3 Ca 381.3 285.5 119.0 82 1 89.2 76.7
Mg 383.0 20.5 15.5 19 5 15.9 16.9
P 7.5 36.5 19.0 - - —
K 25.5 13.0 13.0 8.3 28.6 28.6
9 06 119.0 - — 75.3 - -
Mg 15.5 10.0 10.0 16.5 71.9 71.9
P 9.0 23.3 18.0 - - -
5 K 63.0 25.3 13.0 7 6 27.9 28.6
Ca 393.0 38.0 — 80 3 22.6 -
Mg 31.0 13.7 10.0 12 1 99.6 71.9
P 9.0 27.0 23.0 - - -
6 K 101.0 92.0 13.0 19.9 29.9 9.9
Ca 571.5 76.0 119.0 69.9 96.7 83.0
Mg 62.5 13.7 10.0 15.6 28.9 12.1
P 66.0 78.0 21.0 - - -
K 25.0 19.0 13.0 8.9 18.7 28.6
7 Ca 119.0 57.0 - 79 9 39.7 -
Mg 10.0 10.0 10.0 11 6 91.5 71.9
P 97.0 63.0 31.5 - - -
8 K 38.0 13.0 - 13.0 69.3 -
Ca 119.0 - - 75.9 - -
Mg 10.0 5.0 11.1 35.7 -

 

 

 

 

 

 

 

 

 

 

i ' NH

 

123

SUMMARY OF LABORATORY ANALYSIS DATA

 

 

 

 

 

 

(continued)
Plot Nutrients A B C A B C
Aye lbs/acre Ave percent
Exch. Bases

P 5.0 50.0 25.0 - - -
K 50.5 9.5 38.0 9.9 3.5 11.6
9 Ca 171.5 119.0 119.0 76.7 89.2 67.7
Mg 20.5 10.0 21.0 19.0 12.3 20.8

P 19.0 96.3 30.0 - - -
10 K 76.0 50.7 38.0 9.0 9.5 6.9
Ca 393.0 288.8 229.0 79.1 73.8 80.8
Mg 31.0 28.7 21.0 11.9 16.7 12.3

P 26.0 17.3 28.0 - - -

. K 31.5 13.0 - 10 0 28.6 -

11 Ca - - — - - -
Mg - 10.0 10.0 - 71.9 100.0

P 27.0 61.0 37.0 - - -
12 K 19.0 21.3 13.0 6.9 76.7 16.9
Ca 119.0 76.0 119.0 15.0 53 0 32.0
Mg 15.5 10.0 10.0 9.9 83.0 12.1

 

 

 

 

 

 

 

 

 

 

 

 

m‘-9-A§I‘Lo-.\