.3: ‘ .
.dr

. v

r.
.1!
3&1.» A

11.: .4

.24. .. .

 

I. . t
In“: «adh-
,.1 . 39 Yo. xrl.

. r to 1.?

RN

‘40»:

.2 3. K:
.59
«.3 v9.4?

urea-1....

 

3” hfiuau

:«w-‘s;

ma

1 v1.1...
2} . (L27

 

 

c) l

 

 

mm \m xi

3 1293 01

Ty uimlES
8

x; W

 

 

 

 

 

I

 

This is to certify that the
dissertation entitled

COMPRESSION OF THREE SOILS UNDER LONG-TERM
TILLAGE AND WHEEL TRAFFIC

presented by

Moacir de Souza Dias Junior

has been accepted towards fulfillment
of the requirements for

Ph.D. degreein Croggnd Soil Sciences/
8011 Physics

 

 

Date October 14, 1994

MS U is an Affirmed" Action/Equal Opportunity Institution 0- 12771

 

LIBRARY
Michigan State
University

 

 

 

PLACER RETURN aoxmmwombmummm
TOAVOIDFINESMunonorbdonddodm.

DATE DUE DATE DUE DATE DUE 1
, I ;

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU ismmmqfl OMIM
mm:

COMPRESSION OF THREE SOILS UNDER LONG-TERM
TILLAGE AND WHEEL TRAFFIC

By

Moacir de Souza Dias Junior

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Crop and Soil Sciences

1994

ABSTRACT

COMPRESSION OF THREE SOILS UNDER LONG-TERM TILLAGE AND
WHEEL TRAFFIC

By

Moacir de Souza Dias Junior

Extremes in weather during critieal periods, together with a move to
conservation tillage systems, has renewed concerns over soil compaction during field
operations in agricultural soils. This study examined the compressive behavior of
three Michigan soils in response to changes in soil properties induced by tillage and
wheel traffic; proposed a two component model of soil compressibility that accounts
for stress history, and presented a spreadsheet procedure for estimation of the
preconsolidation pressure (0,). Intact soil cores were equilibrated at four soil water
contents and subjected to uniaxial confined compression tests over the range 25-1600
kPa applied stress. Near-surface penetrometer measurements were made weekly in
1993 on the Capac soil. In general, no-tillage (NT) shifted the compression curves
due to higher bulk densities (pg), increased the preconsolidation pressure (0,) in the
Capac and Kalamazoo soils but not in the Misteguay, and had little effect on the
compression index (m) in any of the soils. The unconfined strength (US) of the
Capac soil confirmed laboratory measurements of 0,, with NT and wheel tracked soil
having higher US than conventional plow. Wheel traffic also shifted the position of
the compression curves, increased 0,, and decreased m. No-tillage had some effect
but wheel traffic did more to decrease the susceptibility of these soils to further

compaction by decreasing m and increasing 0,. The stress history model relates 0, as

a function of water content (0.) as 0, = 10 “ * W. The virgin compression model
takesthe form pm = p, + mlog (am/0), where0isapplied stress, and mis the
compression index modeled as a function of 0. as m = a 4- b0. + c0}. The stress
history model predicted reasonably well 0, (R2 = 0.84 and 0.86) and the log", 0,

(R2 = 0.78 and 0.89) for the data reported in the literature. Field unconfined stress
(US) measurements followed the stress history model and were linearly related to 0,
(R2 > 0.98). A combined spreadsheet procedure was proposed to estimate 0, for
unsaturated soil conditions that compared well to published results and provided a fast

and reliable estimation of 0,.

To M6nica, Neto and Rod
and

in memory of my mother Iracy.

iv

ACKNOWLEDGMENTS

I give my sincerest appreciation and thanks to my advisor Dr. Francis Pierce,
for his guidance, encouragement, assistance in the preparation of this manuscript and
for his constant friendship. Appreciation is also extended to Dr. Thomas Wolff, Dr.
James Crum and Dr. Alvin Smucker for serving on my guidance committee. I thank
Dr. Thomas Wolff for his help and the use of laboratory. In addition, I would like
to thank the Brazilian government through MEC - CAPES for the financial support
and the DCS - ESAL for being part of the process.

Acknowledgment is also made to Luiz, Humberto, and Joana Dias, Gaye
Burpee, Kathy Emmenecker, Jonathan Landeck, Jose Oswaldo, Marcos and Leadir
Fries, Reimar Carlesso and Everson for my beginning. I thank Brian Long for his
technical assistance on the research. Appreciation is also extend to Lima, Cora and
Chuanguo.

Finally, I would like to thank my wife M6nica and my children Moacir and
Rodrigo who have always supported and encouraged me to complete this degree and

to whom I am grateful for their love and understanding.

TABLE OF CONTENTS

LIST OF TABLES ..................................... viii
LIST OF FIGURES ...................................... ix
LIST OF SYMBOLS ..................................... xii
INTRODUCTION ........................................ 1
List of References .................................... 3
LITERATURE REVIEW ................................... 5
Soil compaction process ................................ 5
Modeling soil compaction .............................. 8
Methods to determine the preconsolidation pressure .............. 10
List of references .................................. 14

CHAPTER 1. Soil compressibility of three glacial soils in response to tillage and

wheel traffic ...................................... 27
Abstract ......................................... 27
Introduction ..................................... 29
Material and Methods ................................ 30
Soils ...................................... 30
Soil sampling ................................ 31
Laboratory compression measurements ................. 32
Field unconfined strength measurements ................ 33
Statistics ................................... 34
Results and Discussion ............................... 34
Initial soil properties ............................. 34
Soil compression curves ........................... 36
Field measurements ............................. 39
Summary ........................................ 40
List of References ................................... 42

vi

CHAPTER 2. Accounting for stress history in modeling soil compaction ..... 59

Abstract ......................................... 59
Introduction ...................................... 61
Material and Methods ................................ 63
Model development ............................. 63

Model validation ............................... 65
Results and Discussion ................................ 66
Model validation ............................... 66
Conclusions ...................................... 68
List of References ................................... 70

CHAPTER 3. A spreadsheet procedure for estimating preconsolidation pressure

from soil compression curves ............................ 83
Abstract ......................................... 83
Introduction ...................................... 85
Review of current methods ......................... 87
Material and Methods ................................ 89
Spreadsheet procedure ............................ 89
Results and Discussion ................................ 91
Spreadsheet procedure overview ...................... 94
Conclusions ...................................... 94
List of References ................................... 96
SUMMARY AND CONCLUSIONS ........................... 108

APPENDIX 1. Cells of the suggested spreadsheet prowdure for estimation of the
preconsolidation pressure from soil compression curves ........... 111

APPENDIX 2. Computer screen and cells of the free flow spreadsheet for
computation of the compressibilty test ...................... 112

vii

LIST OF TABLES

LITERATURE REVIEW
Table 1. Relationship between soil properties used to assess soil compaction . . . 22
CHAPTER 1

Table 1. Soil properties of the Capac loam, Kalamazoo loam and Misteguay
silty clay soils averaged across treatments .................. 46

Table 2. Bulk density prior to compression test for the Capac learn, Kalamazoo
loam soils. ..................................... 47

Table 3. Coefficients of the regression of bulk density (p.,) on soil water content
(0.) prior to compression test for the Misteguay silty clay using the
regression model (p, = a + b 0.) ....................... 48

Table 4. Comparison of regression equations of the form (rn = a + b9.I + cog)
for compression index (m) and soil water content (9.) for Capac
loam, and Misteguay silty clay ......................... 49

Table 5. Comparison of regression equations of the form (0, = 10" + " W) for
preconsolidation pressure (0,) and soil water content ( 0.) for Capac
loam, Kalamazoo loam, and Misteguay silty clay ............. 50
CHAPTER 3
Table 1. Preconsolidation pressure (0,) obtained from current literature and
using method 1 through 5 for saturated and unsaturated soil
conditions ..................................... 99

Table 2. Regression equations of preconsolidation pressure (0,) from current
literature and as determined by methods 1 through 5 ........... 101

viii

LIST OF FIGURES

CHAPTERl

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Soil water characteristic curves for the Capac loam, Kalamazoo loam,
and Misteguay silty clay soils at 0—3 cm and 15-18 cm depth for the
NTBT treatment ................................. 51

Soil compression curves for the Capac loam, and Misteguay silty clay

soils and normalized compression curves for the Misteguay soils as
affected by water content (0.). (Error bars represent the standard error

of the mean; error bars for some points are masked by symbols due to
very small std error values) .......................... 52

The relationship between the compression index (m) and soil water
content (0.) for the Capac loam for the 0-3 cm and 15-18 cm depths . 53

The relationship between compression index (m) and soil water
content (0.) for the Kalamazoo loam and Misteguay silty clay for
0—3 cm depth ................................... 54

Compression curves at -6 kPa and ~100 kPa matric potential for

Capac loam, Kalamazoo loam, and Misteguay silty clay soils under

CTBT treatments at 0-3 cm depth. (Error bars represent the standard
error of the mean; error bars for some points are masked by symbols

due to small std error values) ......................... 55

Compression curves at ~100 kPa matric potential (0.) for Capac loam,
Kalamazoo loam, and Misteguay silty clay soils under different tillage
treatments at 0-3 cm depth. (Error bars represent the standard error

of the mean; error bars for some points are masked by symbols due to
small std error) .................................. 56

The relationship between the preconsolidation pressure (0,) and soil

water content (0.) for the Capac loam, Kalamazoo loam , and

Misteguay silty clay soils for the 0-3 cm depth for different tillage

and traffic positions ............................... 57

ix

Figure 8.

Unconfined strength (US) as a function of soil water content (0.) for
the Capac loam for the 0-3 cm depth for different tillage and traffic
positions ...................................... 58

CHAPTERZ

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5 .

Figure 6.

Figure 7.

The secondary compression, rebound, and virgin compression

components of a typical soil compression curve illustrating the

position of the preconsolidation pressure (0,), the critical stress (0,),

the compression index (m), and the shift down and to the left of the

curve with increasing soil water content (0.). The dashed line

represents a field compression curve constructed from the

proposed model .................................. 74

The stress history model (a) expressing preconsolidation pressure (0,)

as a function of soil water content (0.); and the virgin compression

model (b) expressing bulk density (p,) as a function of applied stress

(0) of the 0—3 cm depth for the Capac loam for four different 0. . . . . 75

Soil compression curves expressing bulk density (p,) as a function of
applied stress (0) for the 0-3 cm depth for the Capac loam at four

different 0.. The dashed line represents the line of the regression of
preconsolidation pressure (0,) as a function of soil water content (0.) . 76

Preconsolidation pressure (0,) (from Kassa, 1992 and Reinert, 1990)

and critical stress (0,) (from Kassa, 1992 and Larson and Gupta, 1980)

as a function of soil water content (0.) compared with 0, predicted

from the stress history models obtained from the 0—3 cm depth of the
Capac loam, Kalamazoo loam, and Misteguay silty clay for the
conventionlly tilled treatment ......................... 77

Predicted and measured (Reinert, 1990) values of preconsolidation

pressure (0,) using the stress history model for the 0—3 cm depth of
Kalamazoo loam (5a) and the stress history model for the 0-3 cm

depth of the Capac loam to compare with measurements of

Kassa (1992) (5b). The stress history models used were for

conventionlly tilled treatment ......................... 78

The relationship between critical stress (0,) and preconsolidation
pressure (0,) from Kassa (1992) ........................ 79

The relationship between critical stress (0,) measured by Kassa (1992)
and Larson and Gupta (1980) and preconsolidation pressure (0,)
predicted using the stress history model derived from the 0-3 cm depth

Figure 8.

Figure 9.

of Capac loam when conventionlly tilled .................. 80

Unconfined strength (US) or predicted preconsolidation pressure (0,)

as related to soil water content (0.) in (a) and 0, as predicted US from

(b) using data from 0 - 3 cm depth of the Capac loam in no—till-track
(NT!) and conventionally-tilled-between-track (CTBT) treatments . . . 81

The relationship between critieal strength at which root elongation
ceases, (as predicted from Gerard et al., 1982) and predicted
preconsolidation pressure (0,) for 0. = 0.10 kg kg“ or 4:. at -1.5 MPa
matric potentials for the 0 - 3 cm depth of the Capac loam, Kalamazoo
loam, and Misteguay silty clay soils for conventionally tilled between

CHAPTER3

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Illustration of published methods for determination of the
preconsolidation pressure (0,) for soil compression curves ....... 102

Illustration of methods 1 through 4 for determination of the
preconsolidation pressure (0,) for soil compression curves ....... 103

The effect of water content on the soil compression curves for a Capac
loam soil ..................................... 104

Regression of preconsolidation pressure determined by the Casagrande
(1936) procedure (0,c) on preconsolidation pressure estimated by

methods 1 through 5 (0,“) for 288 compression curves from three soil
series in Michigan ............................... 105

Regression of preconsolidation pressure determined by the Casagrande
(1936) procedure (0,C) on preconsolidation pressure estimated by
combinations of methods 1 and 5 with methods 2 and 3 (0,“) for 288
compression curves from three soil series in Michigan ......... 106

Reproduction of the computer screen of the spreadsheet for

determination of the preconsolidation pressure for soil compression
curves ...................................... 107

xi

a
A, B, C, B

All, BR! CH

AWC

[C]
CI

0

'6

U U U
.g

0

LIST OF SYNIBOLS

Vertieal intercept on q-axis

Soil parameters

Soil parameters determined from load compaction data
during hydrostatic load, 1,. = 0

Available water eapacity

Constant

Cohesion

Stress strain matrix

Cone index

Initial cone index

Clay content

Ratio between 03 and 0:

Particle packing density

Maximum limiting packing density

Optimal degree of compaction

Void ratio

Coefficient for the component of natural volummetric

strain due to shearing stress

xii

Fe

Lk

Su

thnssofcmmpmnhwfoux

Dithionite Fe

Parameters for adjusting the compression curve
Humus content

Measures of how rapidly the maximum packing density
.ktmmhwdvfiflrmmnanngpmummc,0
SammmuimwhmMRuxmdmmhmw

Measures of how rapidly the maximum packing density
is attained with increasing pressure, 0

Air capacity

Liquid limit

Pommuy

bkmqwamnmewmmncammhy

hfidflqxnofihi

Compression index

Slope of the secondary compression curve

Effective vertical overburden pressure

Organic C content

Ratio of maximum shear stress and mean normal stress
Smuhamumt

IkgnwcfismMHMon

Undrained shear strength

xiii

<<c:=

:5 *<

N

AZ
or
5i = Palm»;

6is = (5i’1)Pi

........

Resistance to penetration at field capacity
Desired degree of saturation

Slope of the bulk density vs degree of water saturation
curve

Degree of saturation corresponding to p; and 0.,
Minimum pore water pressure

Pore water pressure

Silt content

Volume

Initial volume

Aggregate tensile strength

Yield function for the plastic behavior

Depth of the specific layer

Change in the depth of specific layer

Slope of failure surface

Normalized initial bulk density

Adjusts the compressive curve for differences in the
initial bulk density of each sample

Major principal strain

Minor principal strain

Equal to { a. 8,, a. 8.}7

Natural octahedral normal strain

xiv

a, = AV/V,

a, = 1n (V/V.)

Pb
Par
phi
Pu:
Pm

pbo

Equal to 2..., when the coefficients have been determined
from triaxial test where 01/03 = 1

Equal to 2..., when the coefficients have been determined
from triaxial test where 01/03 > 1

Volumetric strain

Natural volummetric strain

Total volummetric strain

Volummetric strain in the vertical direction
Volummetric strain in the radial direction

Volummetric strain in the tangential direction

Angle of internal friction

Volummetric soil water content

Water content by weight

......... Initial water content by weight

Optimum water content

Bulk density

Final bulk density

Initial bulk density

Bulk density at a know stress 0.,

Mean of initial bulk density

Bulk density resulting from previous vehicle loading

Applied stress

XV

{0} ........ Equal to { 0,. 0,, 0,, 0..}T

0l ........ Major principal stress

03 ........ Minor principal stress

0c ........ Critical stress

0, ........ Final stress

0, ........ Confining strain

0, ........ Initial stress

0., ........ Reference applied stress = 98 kPa

0. ........ Mean normal stress

0, ........ Normalized stress at u. = 1

0,. = (0, + 0, + 0,)l3 ..... Mean normal stress or octahedral normal stress or
spherical pressure

0, ........ Preconsolidation pressure

0, ........ Residual stress (soil without preconsolidation pressure
0,=O)

0, ........ Applied stress at u = 0

0, ........ Stress at x axis

01 ........ Stress at z axis

0’ ........ Effective vertical overburden pressure

1.. ........ Maximum shearing stress

1,. ........ Octahedral shearing stress

v ........ Specific volume = total volume / volume of solids

VI.

........ Specific volume at 0, = 100 kPa and 0., = 0.20 kg kg"
........ Angle of internal friction in degrees

......... Mattie potential

xvii

INTRODUCTION

Extremes in weather during critical periods, together with a move to
conservation tillage systems, has renewed concerns over soil compaction during field
operations in agricultural soils. Consequently, considerable research has been .
conducted (Barnes et al., 1971; Pidgeon and Soane, 1977; Bauder et al., 1981;
Voorhees, 1983; Gupta et al., 1985; Voorhees et al., 1986; Hakansson et al., 1988;
Larson et al., 1989; Hill and Meza-Montalvo, 1990; Bicki and Simens, 1991; Lebert
and Horn, 1991) to obtain quantitative measurements of changes in soil physical
properties eaused by tillage operations and wheel traffic that would affect plant
development and food production.

Field operations done when soil is too wet for tillage can lead to stress
application that exceeds the soil strength, resulting in unrecoverable deformations.
Farmers, however, have reported that soil managed under no-tillage are more easily
trafficked under high moisture conditions than tilled soils. This could be an important
advantage, particularly in the harvest of crops in wet seasons. However, the exact
condition that defines when a soil is too wet to till or traffic still remains to be
determined. Therefore, not only is the management system an important economical
factor in industrialized agriculture (Bouma, 1984), but knowing when a soil is too wet

for agricultural operations is critical. While moisture conditions and stress history

1

2
primarily govern soil compressive behavior, there are no studies that had quantified
the effects of drying on soil compressibility (McNabb and Boersma, 1993).

In this study, a stress history approach was developed in order to improve
understand of the soil compaction process. The purpose of the first study was to
examine how changes in soil properties induced by tillage and wheel traffic impacted
the compressive behavior of different soils and the extent to which no-tillage and/or
wheel traffic improves traffieability under high soil moisture conditions. The second
study proposes a two component model of soil compressibility, consisting of a stress
history submodel that describes the load carrying capacity of the soil in terms of the
preconsolidation pressure and a classical virgin compression submodel that describes
the plastic, non-recoverable deformation in terms of bulk density and applied stress,
with both submodels as a function of the soil water content. Also, a field based soil
compression curve was proposed based on field measurements of unconfined strength
and water content, which were related to laboratory measurements of preconsolidation
pressure, critical stress, and compression index. Finally, a spreadsheet procedure was
developed to estimate the preconsolidation pressure from uniaxial compression test for

unsaturated soil conditions which was used in the proposed model as a measure of the

soil carrying capacity.

LIST OF REFERENCES

Barnes, K.K, W.M. Carleton, H.M. Taylor, R.I. Throckmorton, and GE.
Vanden Berg. 1971. Compaction of agricultural soils. ASAE. Monogr.,
St. Joseph, MI.

Bauder, J.W., G.W. Randall, and LB. Swan. 1981. Effect of four continuous
tillage systems on mechanical impedance of a clay loam soil. Soil Sci.
Soc. Am. J. 45:802-806.

Bicki, T.J., and J .C. Siemens. 1991. Crop response to wheel traffic soil
compaction. Trans. ASAE 34:909-913.

Bouma, J. 1984. Estimating moisture-related land qualities for land evaluation.
p. 61-75. In Miller, F.P., E.L. Skidmore, D.T. Lewis, and D.M.
Bandel (eds.) Land use planning techniques and policies. SSSA Spec.
Publ. 12. ASA, CSSA, and SSSA, Madison, WI.

Gupta, S.C., A. Hadas, W.B. Voorhees, D. Wolf, W.B. Larson, and EC.
Schneider. 1985 . Development of quids for estimating the ease of
compaction of world soils. Research Report, Binational Agric. Res.
Development, Bet Dagan, Israel, University of Minnesota, USA.

Hikansson, 1., W.B. Voorhees, and H. Riley. 1988. Vehicle and wheel factors
influencing soil compaction and crop response in different traffic
regimes. Soil Tillage Res. 11:239-282.

Hill, R.L., and M. Meza-Montalvo. 1990. Long-term wheel traffic effects on
soil physical properties under different tillage systems. Soil Sci. Soc.
Am. J. 54:865-870.

Larson, W.B., G.R. Blake, R.R. Allmaras, W.B. Voorhees, and S.C. Gupta.
1989. Mechanics and related processes in structured agricultural soils.
NATO Applied Science 172. Kluwer Academic Publishers, The
Netherlands.

Lebert, M., and R. Horn. 1991. A method to predict the mechanical strength

3

4

of agricultural soils. Soil Tillage Res. 19:275-286.

McNabb, DH, and L. Boersma. 1993. Evaluation of the relationship between
compressibility and shear strength of Andissols. Soil Sci. Soc. Am. J.
57:923-929.

Pidgeon, J.D., and B.D. Soane. 1977. Effects of tillage and direct drilling on
soil properties during the growing season in a long-term barley mono-
culture system. J. Agric. Sci. 88:431-442.

Voorhees, W.B. 1983. Relative effectiveness of tillage and natural forces in
alleviating wheel-induced soil compaction. Soil Sci. Soc. Am. J.
47:129-133.

Voorhees, W.B., W.W. Nelson, and G.W. Randall. 1986. Extend and
persistence of subsoil compaction caused by heavy axle loads. Soil Sci.
Soc. Am. J. 50:428-433.

LITERATURE REVIEW

SOIL COMPACTION PROCESS

Soil compaction refers to the compression of unsaturated soils during which
there is an increase in soil density with a reduction in soil volume (Gupta and
Allmaras, 1987; Gupta et al., 1989). Research has clearly shown the effect of soil
compaction on soil physical properties (Barnes et al., 1971; Gupta et al., 1985;
Larson et al., 1989; Binger and Wells, 1992). Soil compaction increases bulk density
and soil strength (Trouse, 1971; Taylor, 1971; Hillel, 1982; Lebert et al., 1989;
Wagger and Denton, 1989; Hill and Meza-Montalvo, 1990; Lebert and Horn, 1991),
and decreases total porosity, size and continuity of the pores (Warkentin, 1971;
Hillel, 1982; Smucker and Erickson, 1989). Significant reductions occur mainly in
the volume of large pores, while small pores remain unaffected (Hillel, 1982). Soil
compaction may have beneficial or adverse effects (Parish, 1971; Gupta and
Allmaras, 1987; Smucker and Erickson, 1989; Raghavan et al., 1990). Beneficial
effects have been attributed to improved seed soil contact (Smucker and Erickson,
1989) and increased available water in dry years (Raghavan and McKyes, 1983).
However, excessive soil compaction can limit nutrient uptake, water infiltration and
redistribution, gas exchange, and root development (S mucker and Erickson, 1989;

Bicki and Siemens, 1991) resulting in decreased yields, increased erosion and

5

6
increased power requirements for tillage (Soane, 1990).

Soil compaction, by definition, refers to a compression of unsaturated soil due
to an applied external stress, that results in a decrease in soil volume. The ease with
which unsaturated soil decreases in volume when subjected to a mechanical stress is
called soil compressibility (Gupta and Allmaras, 1987). The compressibility behavior
ofasoilhasbeendescribedasafunctionoftheextemalandinternal soil factors
(Lebert and Horn, 1991). Soil external factors are characterized by the kind of load
(Koolen and Kuispers, 1983; Horn, 1988; Raghavan et al., 1990), while soil internal
factors are influenced by stress history (Harris, 1971; Horn, 1988; Gupta et al., 1989;
Reinert, 1990), water content (Gupta et al., 1985; Bailey et al., 1986), soil texture
(Gupta etal., 1985; Horn, 1988; McBride, 1989), soil structure (Dexter and Tanner,
1974; Horn, 1988), and initial bulk density (Gupta et al., 1985; Culley and Larson,
1987; Reinert, 1990).

Under dry conditions, soil strength may be great enough to support loads and
soil compaction may be not significant (Trouse, 1971; Taylor, 1971; Larson and
Allmaras, 1971). However, any compaction is detrimental to crop yield under wet
conditions (Swan et al., 1987) and could cause yield reduction (Negi et al., 1980;
Carter, 1985; Gameda et al., 1985; Negi et al., 1990; Bicki and Siemens, 1991). In
areas with a short growing season, field operations are carried out as soon as the soils
are considered trafficable, however, under such conditions the soils are probably still
too wet to be trafficable (HAkansson et al., 1988) and traffic often leads to

unrecoverable soil deformation. In contrast, farmers have indicated that soil managed

7

under no—tillage are more easily trafficked under high moisture conditions than tilled
soils. This could be an important advantage, particularly in the sowing and
harvesting of crops in wet seasons. This may be explained by the fact that no-tilled
soils and wheel-traffic increases bulk density and soil strength greater than 50% than
conventionally tilled soils (Hill and Meza-Montalvo, 1990) at moisture conditions at
saturation and slightly above and below field capacity. In addition, Soane et a1. ,
(1982) suggested that a no-tilled soil becomes precompacted and may have acquired
sufficient soil strength to carry traffic without further compaction occurring. In spite
of these observations, the necessity of quantification of the effect of drying on the
compressibility of soils still remains to be determined (McNabb and Boersma, 1993).
Therefore, while the stress history of a soil is greatly affected by the drying process,
there are few studies that have quantified the effects of long-term no-tillage or drying
on soil compressibility. Thus, there are little quantitative data to support the
observation of increased trafficability of soil managed under no-till.

The persistence of soil compaction beyond the current crop caused by previous
traffic have been reported by several researchers (Smith et a1. , 1969; Black et a1. ,
1976; Voorhees, 1977; Voorhees et al., 1978; Pollard and Elliot, 1978; Logsdon et
a1. , 1992). Some of these studies showed the effects of compaction are only
temporarily harmful, however, in the majority of cases, little or no change in the
persistence of soil compaction was observed. Therefore, restoration of soil

compaction, if possible, is costly and time consuming.

MODELING SOIL COMPACTION

The critical concern with soil compaction is to determine when the soil is too
wet to till or traffic and what level of damage will occur to the soil when applied
stresses exceed its carrying capacity. Thus, a soil is too wet at any water content if
plastic deformation occurs. While much is known about the compaction process
(Barnes et al., 1971; Gupta and Allmaras, 1987; and Gupta et al., 1989), there are no
studies that had quantified the effects of drying on soil compressibility (McNabb and
Boerma, 1993), particularly under field conditions. The emphasis on modeling soil
compaction has been focused on the virgin compression curve which, by definition,
defines plastic, unrecoverable deformation, and is generally well described (Larson
and Gupta, 1980; Gupta et al., 1985; Horn, 1989). However, it is the region of
elastic, recoverable deformation (the secondary compression curve) that defines when
a soil can be tilled or trafficked. It is this component of the soil compression curve
that defines the stress history of soil and it not been modeled. Thus, a model that
predicts the maximum stress that a soil can withstand over a range of water contents
without causing soil compaction is needed. This would answer the question whether a
soil can be tilled or trafficked without soil damage.

In order to assess the susceptibility of soils to compaction, the relationship
between compaction and soil properties must be determined. A summary of the
relationship between soil properties used to assess soil compaction is presented in
Table 1. These relationships were obtained using disturbed soil samples (Bailey and

VandenBerg, 1968; Larson etal., 1980; Larson and Gupta, 1980; Grisso et al., 1987;

9

Bailey and Johnson, 1989; O’Sullivan, 1992), and undisturbed soil samples (Smith,
1985; Reinert, 1990; Lebert and Horn, 1991; McNabb and Boersma, 1993). Also
different types of tests, such as uniaxial compression test (Larson et al., 1930; Larson
and Gupta, 1980; Reinert, 1990; O’Sullivan, 1992) and triaxial (Bailey and
VandenBerg, 1968; Bailey et al., 1986; Bailey and Johnson; 1989, Grisso, 1987)
were used with saturated soil samples (Macwa and Boerma, 1993) and with
different water contents (Bailey and VandenBerg, 1968; Dexter and Tanner, 1973;
Larson and Gupta, 1980; Larson et al., 1980; Reinert, 1990; Lebert and Horn, 1991;
O’Sullivan, 1992) to obtain those relationship. Thus, there is no agreement upon
which soil properties should be used in order to predict soil compaction.

In general, five different approaches have been used as the basis for modeling
the compression behavior of the soil: (1) the virgin compression curve (Soehne, 1958;
Bailey and VandenBerg, 1968; Bowen, 1975; Larson et al.,1980; Lebert and Horn,
1991; Binger and Well, 1992), (2) critical stress (Larson and Gupta, 1980; Gupta and
Larson, 1982; Gupta et a1. 1985); (3) the relationship between strain and applied
stress during triaxial tests (Bailey etal., 1984; Bailey et al., 1985; Bailey et al.,
1986; Grisso eta1.,1987; Bailey and Johnson, 1989); (4) finite element analysis
(Perumpral et a1, 1971; Colleman and Perumpral, 1974; Pollock, Jr. et a1. 1986;
Gassman et al., 1989; Raper and Erbach, 1990 a; Raper and Erbach, 1990 b); and (5)
generalized curve fitting techniques (Blackwell and Soane, 1981; Howard et a1; 1981;
Leeson and Campbell, 1983; Angers et a1, 1987, Lebert etal., 1989; Canarache,

1991; Lebert and Horn, 1991) (Table 1). However, none of these models account for

10

the stress history of the soil, although Lebert et al. (1989), Reinert (1990) and Lebert
and Horn (1991), predict the preconsolidation pressure (0,) from soil properties. The
diminished importance of stress history in current models may be related to the fact
that compression tests are usually performed on disturbed soil samples and at
relatively high soil water contents, both of which would tend to mask the stress
history of a soil.

The 0, is an indication of the maximum previously applied stress sustained by
a soil (Holtz and Kovacs, 1981) and defines the limit of elastic deformation in the soil
compression curve. Thus, in agriculture application of stress greater than the highest
previously applied stress should be avoided (Gupta et a1, 1989; Lebert and Horn,
1991) in order to avoid unrecoverable soil deformations. Therefore, 0, is more likely

to be the maximum stress applied to a soil to prevent further soil compaction.

METHODS TO DETERMINE THE PRECONSOLIDATION PRESSURE

A change in the stress acting on a soil will result in some defamation until a
new equilibrium is reached. These deformations are relatively small and recoverable
during secondary compression and unrecoverable during primary compression of the
soil (Stone and Larson, 1980; Gupta eta1.,1989; Lebert and Horn, 1991). The
preconsolidation pressure has been used to divide the compression curve into regions
of small, elastic and recoverable deformations (secondary compression curve) and in
regions of plastic and unrecoverable deformations (virgin compression curve) (Holtz

and Kovacs, 1981; Jamiolkowski et al., 1985). Thus, additional soil compaction only

11

occurs in the virgin compression curve (Gupta et al., 1989; Lebert and Horn, 1991).
Hence, a consistent, fast, repeatable and reliable method for determination of the
preconsolidation pressure is often of considerable importance from the point of view
of avoiding and predicting soil compaction.

Several methods have been proposed for determining the preconsolidation
pressure from laboratory tests. The Casagrande (1936) method involves selecting the
point of minimum radius of curvature. This is accomplished by drawing horizontal
and tangent lines at this point and bisecting the angle between them, then extending
the straight line portion of the virgin compression curve until it intersects the bisector
of the angle. The pressure corresponding to this point of intersection is the estimated
preconsolidation pressure.

Burmister (1951) proposed a procedure in which the unloading-reloading stress
cycle defines the slope of a typical unloading curve and the form and size of the
characteristic triangle on a semi logarithmic plotting of the curve. By shifting the
unloading curve upward parallel to itself to a point where a geometrically similar
triangle of the same vertical intercept is found, the preconsolidation pressure can be
determined. The preconsolidation pressure is equal to the position of the vertical leg.

Schmertmann (1955) suggested a procedure in which a horizontal line is drawn
parallel to the log of applied stress from the initial void ratio to the existing vertical
overburden pressure. A line parallel to the rebound-reload curve is drawn through
the vertical overburden pressure, and the laboratory initial virgin compression curve is

extended until it intersects either the initial void-ratio or the rebound line. The

12
intersection point is defined as the preconsolidation pressure.

Sallfors (1975, as cited by Larson, 1986) suggested a method in which the two
straight parts of the stress-strain curve are extended and intersected. An isosceles
triangle is inscribed between the lines and the stress-strain curves. The intersection
point between the base of the triangle and the upper line represents the
preconsolidation pressure. '

Anderson and Lukas (1981) predict the preconsolidation pressure (0,) from the
undrained shear strength (Su) and the effective vertical overburden pressure (0’):

0, = Su/(Su/0’)

Culley and Larson (1987) used a statistical procedure to estimate the
preconsolidation pressure. First, a least square regression was determined considering
that all points lay on the virgin compression curve. Next, the compression curve was
divided into two regions assuming an initial estimate of preconsolidation pressure of
15 kPa. Regression equations for each region was them developed and a combined
sums of square calculated. The estimate preconsolidatiOn pressure was then
incrementally increased by 5 kPa and the statistics recalculated until the lowest
residual sums of squares was achieved.

Jose et a1. (1989) used a log-log method in which the applied pressure and
corresponding void ratio are plotted in logarithmic scale for each segment of the
curve. The preconsolidation pressure is assumed to be equal to the applied stress at
the intersection of these two distinct lines. The authors did not reveal their criteria

for choosing which points were included in the calculation of the two lines.

13

Lebert and Horn (1991) estimated the preconsolidation pressure as the
intersection of the regression lines fitted through the secondary compression curve and
the virgin compression curve. The authors did not reveal their criteria for choosing
which points were included in the calculation of the two lines.

Therefore, there are no agreed upon methods for determining the
preconsolidation pressure. However, according to Leonards (1962) the earliest and
most widely used procedure to determine the preconsolidation pressure is the

Casagrande (1936) procedure.

LIST OF REFERENCES

Amir, 1., G.S.V. Raghavan, E. McKyes, and R.S. Broughton. 1976. Soil
compaction as a function of contact pressure and soil moisture content.
Can. Agric. Eng. 18:54-57.

Anderson, T.C., and R.G. Lukas. 1981. Preconsolidation pressure predicted
using Su/p’ ratio. p. 502-515. In Yong, RN. and EC. Townsend
(eds.) Laboratory shear strength of soil. Symposium ASTM. Special
Technical Publication 740. Chicago, Ill, 25 June 1980. Philadelphia,
Pa.

Angers, D.A., B.D. Kay, and RH. Groenevelt. 1987. Compaction
characteristics of a soil crepped to corn and bromegrass. Soil Sci. Soc.
Am. J. 51:779-783.

Bailey, A.C., and CE. Johnson. .1989. A soil compaction model for
cylindrical stress state. Trans. ASAE 32:822-825.

Bailey, A.C., C.E. Johnson and R.L. Schafer. 1984. Hydrostatic compaction
of agricultural soils. Trans. ASAE 27:952-995.

Bailey, A.C., C.E., Johnson, and R.L., Schafer, 1986. A model for
agricultural soil compaction. J. Agric. Eng. Res. 33:257-262.

Bailey, A.C., C.E. Johnson, R.L. Schafer, T.A. Nichols, and R.D. Grisso.
1985. A compaction model for agricultural soil. p. 2:203-212. In
Taction and transport as related to cropping systems. Proc. Int. Conf.
Soil Dynamics. Auburn, AL. June 17-19, 1985. Natl. Tillage Mach.
Lab. and Agric. Eng. Dept., Alabama Exper. Sta., Auburn, AL.

Bailey, A.C., and GE. VandenBerg. 1968. Yielding by compaction and shear
in unsaturated soils. Trans. ASAE 11:307-311,317.

Barnes, K.K., W.M. Carleton, H.M. Taylor, R.I. Throckmorton, and GE.

Vanden Berg. 1971. Compaction of agricultural soils. ASAE. Monogr. ,
St. Joseph, MI.

14

15

Bicki, T.J., and LC. Siemens. 1991. Crop response to wheel traffic soil
compaction. Trans. ASAE 34:909-913.

Binger, R.L., and LG. Wells. 1992. Compact - a reclamation soil compaction
model part 1. model development. Trans. ASAE 35:405-413.

Black, G.R., W.W. Nelson and R.R. Allmaras. 1976. Persistence of subsoil
compaction in a mollisol. Soil Sci. Soc. Am.J. 40:943-948.

Blackwell, P.S., and B.D. Soane. 1981. A method of predicting bulk density
changes in field soils resulting from compaction by agricultural traffic.
J. Soil Sci. 32:51-65.

Bolling, I.H. 1985. How to predict the soil compaction of agricultural tires. p.
936-952. In Taction and transport as related to cropping systems. Int.
Conf. on Soil Dynamics. Auburn, AL. June 17-19,1985. National
Tillage Machinery Laboratory and Agricultural Engineering Dept. ,
Alabama, Experimental Station, Auburn, AL.

Bowen, H.D. 1975 . Simulation of soil compaction under tractor-implement
traffic. ASAE paper No. 75-1569. Am. Soc. Agric. Eng. St. Joseph,
MI.

Brandon, J.R., T. Kuppusamy, and J .V. Perumpral. 1987. A stress-strain
model for agricultural soils. ASAE paper No. 87-1552. Am. Soc.
Agric. Eng. St. Joseph, MI.

Burmister, D. 1951 . The application of controlled test methods in
consolidation testing. p. 83-98. In Fifty-Fourth Annual Meeting of the
ASTM. Symposium on consolidation testing of soils. Special Technical
Publication 126. Atlantic City, NJ. June 18, 1951. Philadelphia, Pa.

Canarache, A. 1991. Factors and indices regarding excessive compactness of
agricultural soils. Soil Tillage Res. 19:145-164.

Carter, L.M. 1985. Wheel traffic is costly. Trans. ASAE 28:430-434

Casagrande, A. 1936. The determination of the pre-consolidation load and its
practical significance. p. 60-64. In Int. Conf. on Soil Mech. and
Found. Eng. Proc. of ICSMFE. Cambridge, Mass. June 22-26, 1936.
vol. 3. Cambridge, Mass.

Colleman, GE, and J.V. Perumpral. 1974. The finite element analysis of
soil compaction. Trans ASAE 17:856-860.

16

Culley, J.L.B., and W.B. Larson. 1987. Susceptibility to compression of a
clay loam Haplaquoll. Soil Sci. Soc. Am. J. 51:562-567.

Dexter, A.R., and D.W. Tanner. 1973. The response of unsaturated soils to
isotropic stress. J. Soil Sci. 24:491-502.

DCXICI. A.R., and D.W. Tanner. 1974. Time dependence of compressibility
for remolded and undisturbed soils. J. Soil Sci. 25353454.

Gameda, S., 6.8. V. Raghavan, R. Theriault, and McKyes. 1985. High axle
load compaction and corn yield. Trans. ASAE 28:1759-1765.

Gassman, P.W., D.C. Erbach, and S.W. Melvin. 1989. Analysis of track and
wheel soil compaction. Trans. ASAE 32:23-29.

Grisso, R.D., C.E. Johnson, and A.C. Bailey. 1987. Soil compaction by
continuous deviatoric stress. Trans. ASAE 30: 1293-1301.

Gupta, S.C., A. Hadas, and R.L. Schafer. 1989. Modeling soil mechanical
behavior during compaction. p. 137-152. In Larson, W.B., G.R. Blake,
R.R. Allmaras, W.B. Voohees, and S.C. Gupta (eds.). Mechanical and
related process in structured agricultural soils. NATO applied sciences
172. Kluwer Academic Publishers, The Netherlands.

Gupta, S.C., A. Hadas, W.B. Voorhees, D. Wolf, W.E. Larson, and EC.
Schneider. 1985. Development of quids for estimating the ease of
compaction of world soils. Research Report, Binational Agric. Res.
Development, Bet Dagan, Israel. University of Minnesota, USA.

Gupta, S.C., and R.R. Allmaras. 1987. Models to access the susceptibility of
soil to excessive compaction. Adv. Soil Sci. 6:65-100.

Gupta, S.C., and WE. Larson. 1982. Modeling soil mechanical behavior
during tillage. p. 151-178. In P.W. Unger, D.M. Van Doren, Jr., F.D.
Whisler, and EL. Skidmore (eds.). Predicting tillage effects on soil
physical properties and process. Spec. Pub. 44. Am. Soc. Agron.
Madison, WI. -

Hakansson, I. 1988. A method for characterizing the state of compactness of
an arable soil. Catena Suppl. 11:101-105.

Hakansson, I., W.B. Voorhees, and H. Riley. 1988. Vehicle and wheel factors
influencing soil compaction and crop response in different traffic
regimes. Soil Tillage Res. 11:239-282.

17

Harris, W.L. 1971. The soil compaction process. p. 9-44 In K.K. Barnes,
W.M. Carleton, H.M. Taylor, R.L. Throckmorton, and GE. Vanden
Berger.(eds.). Compaction of Agricultural Soils. Am. Soc. Agric.
Eng., St. Joseph, MIL ‘

Hill, R.L., and M. Meza-Montalvo. 1990. Long- term wheel traffic effects on
soil physical properties under different tillage systems. Soil Sci. Soc.
Am. J. 54:865-870.

Hillel, D. 1982. Introduction to soil physics. Academic Press, Inc. San Diego,
CA.

Holtz, R.D., and Kovacs, W.D. 1981. An introduction to geotechnical
engineering. Prentice-Hall, Inc., Englewood Cliffs, NJ.

Horn, R. 1988. Compressibility of arable land. Catena Sup. 11:53-71.

Horn, R. 1989. Strength of structured soils to loading - a review of process on
macro and microscale; European aspects. p. 9-22. In W.B. Larson, G.
R. Blake, R. R. Allmaras, W. B. Voorhees, and S. C. Gupta (eds.)
Mechanics and related processes in structured agricultural soils. NATO
applied sciences, vol. 172. Kluwer Academic Publishers, The
Netherlands. ‘

Howard, R.F., MJ. Singer, and G.A. Frantz. 1981. Effects of soil
properties, Water content, and compactive effort on the compaction of
selected California forest and range soils. Soil Sci. Soc. Am. J. 45:231-
236.

Jamiolkowski, M., C.C. Ladd, J.T. Germaine, and R. Lancellotta. 1985. New
development in field and laboratory testing of soils. p. 57-153. In
Publications Committee of XI ICSMFE (ed.). Proc. of the Eleventh Int.
Conf. on Soil Mech. and Found. Eng. San Francisco, CA, 12-16
August 1985. Netherlands.

Johnson, C.E., A.C. Bailey, T.A. Nichols and R.D. Grisso. 1984. Soil
behavior under repeated hydrostatic loading. ASAE paper No.84-1548.
Am. Soc. Agric. Eng. St. Joseph, MI.

Jones, CA. 1983. Effect of soil texture on critical bulk densities for root
growth. Soil Sci. Am. J. 47:1208-1211.

18

Jose, B.T., A. Sridharan, and BM. Abraham. 1989. Log-log method for
determination of preconsolidation pressure. Geotechnical Testing
Journal. 12:230-237.

Koolen, A.I., and Kuipers, H. 1983. Agricultural soil mechanics. Springer
Verlag, Berlin.

Larson, R. 1986. Consolidation of soft soils. Swedish Geotechnical Institute.
Report 29. Linkoping, Swedish.

Larson, W.B., G.R. Blake, R.R. Allmaras, W.B. Voorhees, and S.C. Gupta.
1989. Mechanics and related processes in structured agricultural soils.
NATO Applied Science 172. Kluwer Academic Publishers, The
Netherlands.

Larson, W.B., and R.R. Allmaras. 1971. Management factors and natural
forces as related to compaction. p. 367-427. In k.K Barnes, W.M.
Carleton, H.M. Taylor, R.I. Throckmorton, and GE. Vanden Berg.
Compaction of agricultural soils. ASAE. Monogr., St. Joseph, MI.

Larson, W.B., and S.C. Gupta. 1980. Estimating critical stress in unsaturated
soils from changes in pore water pressure during confined compression.
Soil Sci. Soc. Am. J. 44:1127-1132.

Larson, W.B., S.C. Gupta, and R.A. Useche. 1980. Compression of
agricultural soils from eight soil orders. Soil Sci. Soc. Am. J. 44:450-
457.

Lebert, M., N. Burger, and R. Horn. 1989. Effects of dynamic and static
loading on compaction of structured soils. p. 73-80. In W.B. Larson,
G. R. Blake, R. R. Allmaras, W. B. Voorhees, and S. C. Gupta (eds.)
Mechanics and related processes in structured agricultural soils. NATO
applied sciences, vol. 172. Kluwer Academic Publishers, The
Netherlands.

Lebert, M., and R. Horn. 1991. A method to predict the mechanical strength
of agricultural soils. Soil Tillage Res. 19:275-286.

Leeson, J .J., and DJ. Campbell. 1983. The variation of soil critical state
parameters with water content and its relevance to the compaction of
two agricultural soils. J. Soil Sci. 34:33-44.

Leonards, G.A. 1962. Foundation Engineering. McGraw Hill Book Company,
Inc., NY. .

19

Logsdon, S. D., R. R. Allmaras, W. W. Nelson, and W. B. Voohees. 1992.
Persistence of subsoil compaction from heavy axle loads. Soil Tillage
Res. 23: 95-1 10.

McBride, R.A. 1989. Estimation of density-moismre-stress function from
uniaxial compression on unsaturated structured soils. Soil Tillage Res.
13:383-397.

McNabb, D.H., and L. Boersrna. 1993. Evaluation of the relationship between
compressibility and shear strength of Andisols. Soil Sci. Am. J.
57 :923-929.

Negi, S.C., E. McKyes, F. Taylor, G.S.V. Raghavan. 1980. Crop
performance as affected by traffic and tillage in a clay soil. ASAE
23:1364-1368.

Negi, S. C., G. S. V. Raghavan, E. McKyes, and F. Taylor. 1990. The effect
of compaction and minimum tillage on corn yield sand soil properties.
Trans. ASAE 33: 744-748.

O’Sullivan, M.F. 1992. Uniaxial compaction effects on soil physical properties
in relation to soil type and cultivation. Soil Tillage Res. 24:275-286.

Parish, D.H. 1971. Effects of compaction on nutrient supply to plants. p. 277-
291. In k.K Barnes, W.M. Carleton, H.M. Taylor, R.I. Throckmorton,
and GE. Vanden Berg. Compaction of agricultural soils. ASAE.
Monogr., St. Joseph, MI.

Pollard, F. and LG. Elliott. 1978. The effect of soil compaction and method
of fertilizer placement on the growth of barley using a concrete track
technique. J. Agric. Eng. Res. 23:203-216.

Pollock, Jr. D. J. V. Perumpral, and T. Kuppusamy. 1986. Finite element
analysis of multipass effects of vehicles on soil compaction. Trans.
ASAE 29:45-50.

Prumpral, J.V., LB. Lijedahl, and W.H. Perloff. 1971. The finite element
method for predicting stress distribution and soil defamation under a
tractive device. Trans. ASAE 14:1184-1188.

Raghavan, G.S.V., and E. McKyes.. 1983. Physical and hydraulic
Characteristics in compacted clay soils. J. Terrarnechanics. 19:235-242.

20

Raghavan, G.S.V., P. Alvo, and E. McKyes. 1990. Soil compaction in
agriculture: A review toward managing the problem. Advances in Soil
Sci. 11:1-36.

Raper, R.L., and D.C. Erbach. 1990 a. Prediction of soil stresses using the
finite element method. Trans. ASAE 33:725-730.

Raper, R.L., and D,C. Erbach. 1990 b. Effect of variable linear elastic
parameters on finite element prediction of compaction. Trans. ASAE
33:731-736.

Reinert, DJ. 1990. Soil structural form and stability induced by tillage in a
Typic Hapludalf. Ph.D diss. Michigan State Univ. East Lansing.

Saini, G.R., and TL. Chow. 1984. Compactibility indexes of some
agricultural soils of New Brunswick, Canada. Soil Sci. 137:33-38.

Sillfors, G. 1975. Preconsolidation pressure of soft high plastic clays. Thesis.
Department of Geotechinical Engineering. Gothenburg.

Schmertmann, I.H. 1955. The undisturbed consolidation
behavior of clay. Trans. ASCE 120:1201-1233.

Smith, D.L.O. 1985. Compaction by Wheels: a numerical model for
agricultural soils. J. Soil Sci. 36:621-632.

Smith, S.T., T.C. Stoneman, and C.V. Malcolm. 1969. Cultivation and traffic
hardpans in Swan Valley Vineyards. West. Aust. Dept. Agric., Tech.
Bull. 1.

Smucker, A.J.M., and A.E. Erickson. 1989. Tillage and compactive
modifications of gaseous flow and soil aeration. p. 205-221. In W.B.
Larson, G.R. Blake, R.R. Allmaras, W.B. Voorhees, and S.C. Gupta
(eds.). Mechanics and related processes in structured agricultural soils.
NATO applied sciences, vol. 172. Kluwer Academic Publishers, The
Netherlands. .

Soane, B.D. 1990. The role of organic matter in soil compactibility: a review
of some practical aspects. Soil Tillage Res. 16: 179-201.

Soane, B.D., J.W. DicKson, and D]. Campbell. 1982. Compaction by
agricultural vehicles: a review III. Incidence and control of compaction
in crop production. Soil Tillage Res. 2:3-36.

21

Soehne, W.H. 1958. Fundamentals of pressure distribution and soil
compaction under tractors tires. J. Agric. Eng. 276-291.

Stone, J.A., and WE. Larson. 1980. Rebound of five one-dimensionally
compressed unsaturated granular soils. Soil Sci. Soc. Am. J. 44:819-
822.

Swan, J.B., J .F. Moncrief, and W.B. Voohees. 1987. Soil compaction-causes,
effects, and control. University of Minnesota, Agric. Bull. 3115. St.
Paul, Minnesota.

Taylor, H.M. 1971. Effects of soil strength on seedling emergence, root
growth and crop yield. p. 292-305. In k.K Barnes, W.M. Carleton,
H.M. Taylor, R.I. Throckmorton, and GE. Vanden Berg. Compaction
of agricultural soils. ASAE. Monogr., St. Joseph, MI.

Trouse, JR., A.C. 1971. Present knowledge and need for research. p. 225-
240. In k.K Barnes, W.M. Carleton, H.M. Taylor, R.I. Throckmorton,
and GE. Vanden Berg. Compaction of agricultural soils. ASAE.
Monogr., St. Joseph, MI.

Vanden Berg, GE. 1966. Triaxial measurements of shear strain and
compaction in unsaturated soil. Trans. ASAE 9:460-463,467.

Voorhees, W.B. 1977. Soil compaction: our newest natural resource. Crops
Soils Mag. 29:13-15.

Voorhees, W.B., C.G. Senst, and W.W. Nelson. 1978. Compaction and soil
structure modification by wheel traffic in the Northern Corn Belt. Soil
Sci. Soc. Am. J. 42:344-349.

Wagger, M.G., and HP Denton. 1989. Influence of cover crop and wheel
traffic on soil physics properties in continuous no-till corn. Soil Sci.
Soc. Am. J. 53:1206-1210.

Warkentin, BR 1971. Effects of compaction on content and transmission of
water in soils. p. 126-153. In k.K Barnes, W.M. Carleton, H.M.
Taylor, R.I. Throckmorton, and GE. Vanden Berg. Compaction of
agricultural soils. ASAE. Monogr., St. Joseph, MI.

Wlodek, S. 1991 . Depth indicators method for determination of bulk density
dynamics. Soil Tillage Res. 19:197-201.

22

Table 1. Relationship between soil properties used to assess soil

compaction.

 

Reference

Relationship

 

Soehne, 1958
VandenBerg, 1966

Bailey & VandenBerg, 1968

Dexter & Tanner, 1973

Colleman & Perumpral, 1974

Bowen, 1975

Amir et al., 1976

Larson et al., 1980

Larson & Gupta, 1980

Blackwell & Soane, 1981

n = m In 0 + n,

p = A +Blog[0..,,(1+ 1'...)]

Up, = mloga+B

l/p, =Alog§+B(r../0.) + C

j‘ = (0,.,2 + 0..,3)"2

0. = (0, + 20,)l3

7.. = (0, - 03)/3

D = D, + B exp (-k0) + C exp (-L0)

D = (p l 2660) [(100-OC) I (100+0.)]

8,1- = (-0.007 + 1.72 R - 15.854R2 +

96.107 R3 - 237.304 R‘ + 213.301 R’)* 103
= - m log 0 + C

p, = 2.65 (1 - n/ 100)

n =A-Bln(0,+ 0)-Cln0

pb=A +Bln(0,+ 0)-Cln0

p, = pbk + S; (Sl - S.) + m log (0/0.)

log 0c = 0. log 0,

m and p, = f (0.)

p,, = 1.166 + 0.252 ln 0...,

Table l (cont’d).

Howard et al., 1981

Gupta & Larson, 1982

Jones, 1983

Leeson & Campbell, 1983

Bailey et al., 1984

23

p, = 1.19 - 0.596 OC - 0.076 LL + 0.0019 s
+ 0.0058 Fe

1.93 - 0.0628 OC - 0.0063 LL

p.
+ 0.0012 s
p, = 3.27 - 0.0231 OC - 0.528 In 0..
- 0.0008 s + 0.0039 Fe
11 = f (0., 0)
criteria of :
critical air-filled porosity,
critical stress for shearing,
aggregates and critical soil
resistance for root growth was superimposed.
p, = 1.52 - 0.00646 Cl
for sandy loam soil
0 = 2.25 - 0.008 0.
for loam soil
0 = 2.28 - 0.011 0.
a, = (A+Ba)(l-ec")
a, = MN, AV = v,-v

Up, = l/pbi - Up. (A + B0) (1 - e“)

Table l (cont’d).

Johnson et al., 1984

Saini et al., 1984

Gupta et al., 1985

Bailey etal., 1985
and

Bailey et al., 1986

Bolling, 1985

Smith, 1985

Pollock, Jr. etal., 1986
Angers et al., 1987
Grisso etal., 1987

Brandon et al., 1987

24

6. = (A + 130) (1 - CXP (-C0))
1n p, = 1n p. - (A + B0) (l-exp (-C0))
p, = 1.2926 - 0.2504 0. + 0.8353 0.2
+ 0.9932 0.3 + 0.1203 F - 0.0330F2
+ 0.0026 F3 + 1.0635 0.F +7.4289 0.1F
+ 12.96350.3F + 0.0984 0.F2
- 0.3842 0.1F2 - 0.1272 0.1F’ +
+ 0.02880,.,F3 - 0.2231 0.2F3
' +0.45880.3F3
p. = f (S. 0)
t, = (A + B0.) (1 -c0'_)
'6‘, = ln (V/Vo)
1n(p,) =ln(p,9-(A + B0.)(1-e‘°‘,)
n = no - 0../0.03 [CI/CDT”
n = n, - (n, - 0.225) / (35C, + 1)(0,,,/12)”2 01
Ad: = 0i - (pt-nu) [(02-00 / (nu-p.01
e, = a, + a, + a.
Y = - 112.2 + 88.9 p,
t... = (a... / a...) (A. +3.0.) (1 - ecu-J3
YF=a+a[(0,+0,)/2]-

'{Kax' 0y) I 2]2 + 03,2}10

Table 1 (cont’d).

Hikansson, 1988

Bailey & Johnson, 1989

Lebert et al., 1989

Raper & Erbach, 1990 a
Raper & Erbach, 1990 b
Reinert, 1990
Canarache, 1991

Lebert & Horn, 1991

Wlodek, 1991

Binger &' Wells, 1992

25

for0<Cl<60%;l<H<1l%
D,,,, = 90.5 - 0.29 C1 + 0.0059 Cl2 - 0.139 H
forO < Cl < 60%
D,In = 86.5 + 0.041 C1
2,, = (A +B0,.)(1-e°°“) +E(r.,/0.,)
lnp, =1np.-(A +B0,.) (1 -e°°‘) +
+E (1.,/0,.)
0, = 2.1592 p, + 0.234 LK + 0.0360 AWC
+ 0.0770 NAWC - 3.426
0, = (3.0975 p, - 0.0475 Cl - 0.0280U -
- 0.9659 log s +0.3369 LK - 0.0268 0
+ 2.1330 log c + 0.0839)2
s, = exp[(A + Ba.) (1 - cam-1
{a} = [C] {a}
0, = -263-2.668 + 32216.,i
log RP = - 4.14 + 0.0858 p, - 0.000347»,2
e = B + m log 0
0, = f (¢. c ,p,, LK, AWC, NAWC, Kf, OC)
9. =pn12/(z + 42)]
Secondary compression curve

p. = p...‘ + 111. log (a/ a.)

26

Table l (cont’d).
Virgin compression curve

Pb=Pu+ST(Sr'Sk)+ml°g(0/at)

O’Sullivan, 1992 v = v, - m 1n (0 / 0,)- b(0. - 0..)
McNabb & Boersma, 1993 1n p, = in 005.6,) - (A + B0 +J6,) (1 - e“)
5i = Psi] Phi"

5c = (5i ‘ 1) Poi

 

CHAPTERI

SOIL COMPRESSIBILITY OF THREE GLACIAL SOILS IN RESPONSE TO

TILLAGE AND WHEEL TRAFFIC

ABSTRACT

Field observations indicate soils managed under no-tillage are more easily
trafficked than tilled soils. This study examined how changes in soil properties
induced by tillage and wheel, traffic impacted the compressive behavior of three
Michigan soils. Intact soil cores, from track and between track positions in
conventional moldboard or chisel plow (CT) and no-tillage (NT) treatments from the
Kalamazoo loam (Fine loamy, mixed, mesic, Typic Hapludalfs), the Capac loam
(Fine loamy, mixed, mesic, Aeric Ochraqualfs), and the Misteguay silty clay (Fine,
mixed (calcareous), mesic, Aerie Haplaquepts) were equilibrated at four soil water
contents and subjected to uniaxial confined compression tests over the range 25-1600
kPa of applied stresses. In general, NT shifted the compression curves down due to
higher bulk densities, increased the preconsolidation pressure (0,) in the Capac loam
and Kalamazoo loam soils but not in the Misteguay, and had little effect on the
compression index (m) in any of the soils. Unconfined strength (US) of the Capac

loam soil confirmed laboratory measurements of 0,, with NT and wheel tracked soil

27

28

having higher US than CT. Wheel traffic also shifted the position of the compression
curves, increased 0,, and decreased m. NT treatment had a small effect but wheel
traffic did more to decrease susceptibility of these soils to further compaction by
decreasing m and increasing 0,. Perceptions of increased trafficability of soils in NT
relates not so much to tillage induced differences in soil physical properties but is

due, primarily, to wheel traffic.

29

INTRODUCTION

In recent years, extremes in weather during critical periods, together with a
move to conservation tillage systems, has renewed concern over soil compaction
during field operations in agricultural soils. Also, farmers have reported that soils
managed under no-tillage (NT) are more easily trafficked under high soil water
content than tilled soils. Therefore, increasing soil strength could be an important
advantage of NT treated soils for trafficability under high soil water content, for
example, at harvest. In general, NT increases bulk density (11,) and soil strength
when compared with conventional tillage (Soane et al. , 1982; Hill and Men-
Montalvo, 1990). However, there are few studies that have quantified the effects of
long-term NT on soil compressibility.

Soil management under NT and conventional tillage (CT) has produced
differences in soil physical properties (Pidgeon and Soane, 1977; Bauder et al., 1981)
and quantitative measurements of those changes have been reported for a number of
soils (Voorhees etal., 1978; Gupta et a1, 1985; Culley and Larson, 1987; Horn,
1988; Johnson et al., 1989; Hill and Meza-Montalvo, 1990; Muller etal., 1990;
Meek et a1. , 1992; Pierce et a1. , 1992; Pierce et a1. , 1994). However, few studies
have considered changes in soil compressibility with changes in soil water content
(Culley and Larson, 1987; Reinert, 1990; Kassa, 1992). Therefore, quantification of
the effect of drying on soil compressibility remains to be determined (McNabb and
Boersma, 1993).

The stress history of a soil greatly affects its compressive behavior (Culley and

3O

Larson, 1987; Harris, 1971; Soane et al., 1982). However, soil compressibility is
strongly regulated by soil water content and the concern over soil damage has focused
mainly on soil behavior at high soil water content. Additionally, most soil
compression measurements have been made on disturbed soil samples and at high soil
water content (Bailey and VandenBerg, 1968; Larson and Gupta, 1980; Larson et al.,
1980; Gupta et al., 1985; Grisso et al., 1987; Bailey and Johnson, 1989; O’Sullivan,
1992). Thus, some if not all may have had the stress history altered by the sieving
process or by the high soil water content at which compression tests were conducted.
The purpose of this study was to examine how changes in soil properties
induced by tillage and wheel traffic impacted the compressive behavior of different
soils and the extent to which no-tillage and/or wheel traffic improves trafficability

under high soil water conditions.

MATERIAL AND METHODS
Soils

Soil from experiments managed under long term NT and CT were sampled to
characterize compressive behavior of three glacial soils in Michigan. Soils used in
this study were: Kalamazoo loam (Fine loamy, mixed, mesic, Typic Hapludalfs)
located at the Kellogg Biological Station (KBS), Hickory Corners, MI; Capac loam
(Fine loamy, mixed, mesic, Aerie Ochraqualfs) located on the Michigan State
University Agronomy Farm, East Lansing, MI; and Misteguay silty clay (fine, mixed

(calcareous), mesic, Aerie Haplaquepts) located near Saginaw, MI. Prior to

31

sampling, the experiments have been managed under NT for 13, 14, and 9 yr,
respectively (Bronson, 1989; Pierce et al., 1994; Martinson, 1993; Xu, 1994).
Conventional tillage consisted of fall moldboard plowing for the Capac loam and
Misteguay silty clay soils and spring chisel plowing for the Kalamazoo loam soil, with
secondary tillage in the spring consisting of one pass of a tandem disk and one pass of

a harrow prior to planting for all three soils.

Soil sampling

For each soil, the NT and CI‘ treatments were sampled in three transects
perpendicular to crop rows, both in track (T) and between track (BT) positions, and at
two depths, 0-3 cm and 15-18 cm. The soils were sampled on the following dates:
the Capac loam on 25 August, 1992, the Kalamazoo loam on 18 August, 1992 and
again on 7 May, 1993, and the Misteguay silty clay on 5 April, 1993, with all spring
sampling occurring prior to any field operations. Four soil cores were taken at each
position to allow for compression measurements at four gravimetric soil water
contents (0.). Intact soil cores (6.35 cm diameter and 2.54 cm length) were sampled
using a metal soil sampler containing rings of polyvinyl chloride (PVC) pipe placed
within a cutting metal device with a bevelled cutting edge. The sampling device was
pushed carefully into the soil using a falling weight. The ring filled with soil was
removed from the metal device and the ends were trimmed to the dimension of the

PVC ring. Soil cores were stored in plastic at 4 °C until compressibility tests were

performed.

32

Disturbed soil samples were obtained near the intact soil cores, air dried and
subjected to standardized test for plastic and liquid limits (Sowers, 1986), particle size
analysis by the pipette method, and sand fractionation by sieving (Day, 1986). Bulk
density was determined as dry soil weight per unit volume of the intact soil cores
(Blake and Hartge, 1986). Total organic C and N were determined by dry
combustion of 5 replicate samples of 50 mg on a Carlo Erba CHN analyzer Model

1104 (Carlo Erba Instruments, Milano, Italy).

Laboratory Compression Measurements

To achieve a range of 0., soil cores were saturated and equilibrated to a matric
potential (0.) equal to -6 kPa and -100 kPa on a ceramic plate inside a pressure
chamber (Klute, 1986). For lower 0., soil cores were first equilibrated at a matric
potential of -100 kPa and then air dried at room temperature until within the desired
0. (0.07 to 0.10 and 0.03 to 0.07 kg kg“ for Kalamazoo loam, 0.08 to 0.14 and 0.03
to 0.06 kg kg“ for Capac loam, and 0.16 to 0.23 and 0.09 to 0.14 kg kg“ for
Misteguay silty clay).

Uniaxial compression tests were conducted according to Bowles (1986), using
a pneumatic Brainard-Kilman consolidometer (2175 West Park Ct. Stone Mountain,
GA). The strain measuring device uses a dial gage reading with 2.54 flm/dIVISIOII.
The loads were applied until 90% of maximum deformation was reached. The 90%
of maximum deformation was determined by drawing a straight line through the data

points in the initial part of the curve obtained when dial readings were plotted versus

33
1/time, until this line intercepts the y axis (dial readings). A second straight line was
drawn from this intersection with all abscissas 1.15 times as large as corresponding
values on the first line. The intersection of this second line and the laboratory curve
is the point corresponding to 90% consolidation (Taylor, 1948 as cited by Holtz and
Kovacs, 1981). After this condition was reached, a new successive stress was
applied. Increasing stresses were applied in succession using an applied stress (0)
sequence of 25, 50, 100, 200, 400, 800, and 1600 kPa. The compression index (m)
was computed as the slope of the virgin compression line plotted as p, versus log 0

(Bradford and Gupta, 1986). The preconsolidation pressure (0,) was estimated by the

Casagrande (1936) procedure.

Field Unconf' med Strength Measurements

Three replications of penetrometer measurements were made in the field
weekly in 1993 over a 9-wk period (May, June, and July) for Capac loam soil, with
measurements in track and between track positions of both tillage treatments. A
pocket penetrometer (Soiltest model CL-700, 2205 Lee Street, Evaston, Illinois) was
pushed into the soil until a reference mark was reached and the reading was recorded.
0. were determined for each penetrometer reading by drying soil at 105 °C for 24

hours.

34

Statistics

Regression equations were performed using the computer program Sigma Plot
1.02 (Iandel Scientific, P.O. Box 7005, San Rafael, CA) for p, prior to compression
tests, 0,, m and 0.. Intercepts and slopes of the regression equations of p, prior to
the compression test and preconsolidation pressure were compared using procedures

of Snedecor and Cochran (1967).

RESULTS AND DISCUSSION

The compressibility of soil is a function of several soil factors, primarily soil
water content, texture, structure, stress history, and initial p, (Culley and Larson,
1987; Gupta and Allmaras, 1987; Gupta et al., 1987; Gupta et al., 1989; Horn, 1988;
Larson et a1. , 1980). Therefore, soil physical properties of the three soils and the
effects of tillage and traffic on the initial conditions prior to compression tests will be

discussed first.

Initial Soil Properties

The Kalamazoo loam and Capac loam had similar particle size distribution,
with clay contents between 90 to 110 g kg“, but the sand size distribution was coarser
in the Kalamazoo loam than the Capac loam with geometric mean diameters 0.076
mm and 0.031 mm, respectively (Table 1). The Capac loam had higher OC and N,
slightly higher consistency limits (Table 1), higher water holding capacity (Figure 1),

and lower p, at both depths (Table 2). p, was similar between tillage systems but was

3 5
higher in the wheel-track than between tracks (Table 2). Differences in soil physical
properties between these two soils were related to differences in OC and associated
differences in soil structure (Reinert, 1990; Pierce et al., 1994). Additionally, the
surface 3 cm of all three soils had higher water holding capacities than the 15-18 cm
depth, indicative of lower bulk density (Table 2) and higher OC.

The Misteguay silty clay had clay contents of 480 to 490 g kg“, with very
little sand (50 g kg“ Table l), and had a high water holding capacity (Figure 1). The
consistency limits of the Misteguay silty clay were more than double the other soils
and the plastic index more than tripled (Table l).

. The initial p, of the Misteguay silty clay, prior to the compression tests,
decreased linearly as 0. increased, although the strength of the regression varied with
tillage and traffic condition (Table 3). This is in contrast with the Capac loam and
Kalamazoo loam soils, for which p, was invariant to 0.. Tillage and wheel traffic
shifted the regression curve either in the slope, in the intercept, or both, in the
Misteguay silty clay soil. Statistical tests comparing the regression lines for different
treatments and depths showed that the regression lines were parallel (equal slopes)
with the exception that CTT had a higher slope than CTBT at 0-3 cm depth (Table 3).
Therefore, shrinkage upon drying in the conventionally tilled Misteguay silty clay soil
was greater in the wheel track. Differences in the intercept of the regression lines are
indicative of soil compaction. In the 0-3 cm depth, the tracked soil (both NTT and
CTDhadahigherpbthanbetween tracked (NTBde CTBT) and theNTBTwas

more compact than CTBT. In the 15-18 cm depth, only the N'IT was initially more

36

compact than the untracked soil (both NTBT and CTBT). Therefore, not only did 0.
effect p, in the Misteguay soil, stress history was also very important in determining

this relationship.

Soil compression curves

Soil compression curves were obtained by plotting p, versus applied stress (0).
The compression curve is comprised of two regions: a region of plastic and
unrecoverable deformation called the virgin compression curve, and a region of small,
elastic and recoverable deformation called the secondary compression curve. The
slope of the virgin compression curve is called the compression index (m). The point
that divides these two regions in a compression curve is the preconsolidation pressure
(0,). These parameters define the soil compression curve and may change with soil
type, initial 0., and management history (Culley and Larson, 1987; Larson et al.,
1988).

Soil water content was the major factor regulating the compressive behavior of
these soils (Figure 2). Larson et a1. (1980) reported that as initial 0. increases, soil
compression curves are generally displaced down and to the left in a parallel manner,
indicating an increase in susceptibility of soil to compaction with increasing 0.. This
shift in the compression curves with increasing 0. was true for both the Capac loam
(Figure 2) and Kalamazoo loam (data not shown). The shift in the compression
curves for the Misteguay silty clay was reversed (Figure 2). This appeared to be

related to moisture effects on initial p, as the Misteguay silty clay soil shrinks upon

37

drying (Table 3). This apparent paradox was resolved when the curves were
normalized (p. at each stress was divided by initial p. prior to compression test).
When normalized, compression curves for the Misteguay silty clay soil conformed to
the same pattern of shifting down and to left as 0. increased.

The virgin compression curves for these soils were not parallel, as reported by
Larson et a1. (1980). The non-parallel nature of the virgin compression curves were
apparently here due to the broad range of 0. measured compared to the moisture
range measured by Larson et a1. (1980). This is consistent with Schmertmann (1955),
who reported that compression curves intersect within a narrow range of void ratio,
with an average estimate of 0.42 of the initial void ratio reasonable for most clays.
For these soils, m was a function of 0., but the form of the relationship varied with
soil type. For the Capac loam and Misteguay silty clay soils, the general relationship
between m and 0. followed

m=a+b0.+c0.2 [l]
with R2 ranging from 0.26 to 0.61, with higher R2 for the 0-15 cm depth (Figure 3
and 4 and Table 4). The m was lower for tracked than between tracked soil in the 0-
3 cm depth but not in the 15-18 cm depth and m... occurred at 0. near the plastic
limit. For the Kalamazoo loam, m decreased linearly with increasing 0. for the 0-3
cm depth regardless of wheel traffic (R2 =0.37), but showed little change with 0. at
the 15-18 cm depth (Figure 4). While the relationships between m and 0. are weak,
and we do not understand why the behavior is different for the Kalamazoo loam, the

change in m with 0. has important implications in predicting the amount of

38

defamation per applied stress that will occur in the virgin compression curve.

These soils, however, exhibited a strong dependence of stress history on 0..
At -6 kPa 0., the compression curves showed little stress history (slight curvature at
low applied stress as indicated by a significant fit of the second order polynomial)
(Figure 5a). This relates to the fact that 0. was near the liquid limit for Capac loam,
Kalamazoo loam, and Misteguay silty clay (0.25, 0.22, and 0.53 kg kg“,
respectively), where the stress bearing capacity is limited. Note that the curves for
the three soils are nearly parallel, indicating a similar deformation for a given applied
stress. At -100 kPa 0., 0. was near the plastic limit for Capac loam, Kalamazoo
loam, and Misteguay silty clay (0.17, 0.15, and 0.26 kg kg“, respectively). The
compression curves clearly show the presence of a stress history (curvature at low
applied stress), although this was less so for the Misteguay silty clay than the other
soils (Figure 5b). Therefore, less deformation is expected at -100 kPa 0. at low
applied stress due to the presence of stress history, but higher deformation at higher
applied stresses due to higher m. In the Capac loam and Misteguay silty clay soils,
CTBT and NTBT treatments exhibited a similar compression behavior at -100 kPa 0.
i.e., the two curves were similar for the Capac loam and parallel for the Misteguay
silty clay (Figure 6). In the Kalamazoo loam, the NTBT had a higher initial p, than
the CTBT but lower defamation (curves cross). The effects of tillage were similar at
other 0.. Thus, although NTBT had the same load carrying capacity (similar 0,) as
CTBT in the Kalamazoo loam, the NTBT had lower defamation than CTBT at

applied stress > 0,. 0., therefore, affects both 0, and m, and thus regulates the

39

shape of the compression curve.

Overall, as illustrated in Figure 7, 0, decreases as a function of increasing 0.,

following the relationship

0, = 10“ " m . [2]
The coefficients varied with soil, tillage, and wheel traffic, with R2 ranging from 0.83
to 0.98 (Table 5). Tillage and wheel traffic influenced the relationships between 0,
and 0. (Table 5). For example, NTT was often different from CTBT but not different
from either CTI‘ or NTBT (Figure 7). In the Capac loam and Kalamazoo loam, the
NTT could sustain a higher stress than the other treatments while in the Misteguay at
high 0., this was true for CIT, although NTT was greater than NTBT. A clear
difference did not exist between wheel track and no wheel track for Misteguay silty
clay soil.

Based on the relationships in Equations [1] and [2], at high soil moisture, 0, is
unimportant when the soil is near the liquid limit and m is moderate, therefore,
defamation is not at a maximum. As the soil drains, 0, increases only slightly, but
since the soil must increasingly support more of the applied stress, m increases and
defamation increases. As further drying takes place, 0, increases exponentially and

the soil can support considerable loads without further defamation.

Field measurements
Field penetrometer measurements for the Capac loam showed that unconfined

strength (US) increased exponentially with decreasing 0. (Figure 8). The fom of this

4O

relationship is consistent with that between 0, and 0. measured in the laboratory
(Figure 7). As was the case with 0,, NTT had the highest US and CTBT the lowest.
The NTBT was intermediate but approximately parallel to CTBT and CTT is the
same as NTBT, which upon careful inspection of Figure 7 and Table 5, is also
consistent with the laboratory measurements. At the plastic limit, the NTT soil
strength values were five times greater than for CTBT. Therefore, field data support

conclusions from laboratory measurements.

SUMMARY

Changes in soil properties induced by tillage and wheel traffic affected the
compressive behavior of these three soils. 0. regulated the shape of the curve while
initial bulk density p, regulated its position. The initial p, of the Misteguay silty clay,
and subsequently the compressive behavior, was greatly affected by 0., and required a
nomalization of the compression curves to fit the generalized relationship of shifts in
soil compression curves with changes in 0.. In general, no-tillage shifted the
compression curves, increased 0, in the Capac loam and Kalamazoo loam soils but not
in the Misteguay, and had little effect on m in any of the soils. No-tillage also had
higher field measured unconfined strength than CT in the Capac loam soil. Wheel
traffic shifted the position of the compression curves, due to their influence on initial
conditions, increased 0,, and decreased m. These shifts would support the notion of
improved trafficability on no-tilled and trafficked soils. No-tillage had some effect

but wheel traffic did more to decrease susceptibility of these soils to further

4 1
compaction by decreasing m and increasing 0,. Specifically, wheel traffic in no-
tillage (NTT) had a higher 0, ill the Capac loam and Kalamazoo loam soil, although
CTT was higher in the Misteguay silty clay soil. The perception of increased
trafficability of soils in no-tillage, as reported by famers, relates not so much to
tillage induced differences in soil physical properties but is primarily due to wheel
traffic effects and the fact that controlled traffic is likely in long-term NT.
Additionally, soils that dry faster would support higher loads earlier. Therefore,
famers must not only consider the adoption of controlled traffic patterns to reduce

overall soil compaction but should focus mainly on the enhanced resistance due to soil

drying.

LIST OF REFERENCE

Bailey, A.C. and CE. Johnson. 1989. A soil compaction model for cylindrical
stress state. Trans. ASAE. 32:822-825.

Bailey, A.C., and GE. VandenBerg. 1968. Yielding by compaction and shear in
unsaturated soils. Trans. ASAE. 11:307-311,317.

Bauder, J .W., G.W. Randall, and LB. Swan. 1981. Effect of four continuous
tillage systems on mechanical impedance of a clay loam soil. Soil Sci.
Soc. Am. J. 45:802-806.

Blake, G.R. and K.H. Hartge. 1986. Bulk density. p. 363-375. In A. Klute
(ed.) Methods of soil analysis. Part 1. 2nd ed. Agron. Monogr. 9.
ASA, Madison, WI.

Blackwell, P.S., and B.D. Soane. 1981. A method of predicting bulk density
changes in field soils resulting from compaction by agricultural traffic.
J. Soil Sci. 32:51-65.

Bradford, J .M and S.C. Gupta. 1986. Compressibility. p. 479-492. In A.
Klute (ed.) Methods of soil analysis. Part 1. 2nd ed. Agron. Monogr.
9. ASA, Madison, WI.

Bowles, LA. 1986. Engineering properties of soils and their measurements.
Third edition. McGraw-Hill Book Company, Inc., NY.

Bronson, J .A. 1989. The effect of tillage system on corn production and soil
properties on a Kalamazoo loam. MS Thesis. Michigan State
University. East Lansing.

Casagrande, A. 1936. The detemination of the pre-cansolidation load and its
practical significance. 3:60-64. In Int. Conf. on Soil Mech. and
Foundation Eng. Proc. of the First ICSMFE. Cambridge, Mass. June
22-26,1936. Cambridge, Mass.

42

43

Culley, J.L.B., and W.B. Larson. 1987. Susceptibility to compression of a
clay loam Haplaquoll. Soil Sci. Soc. Am. J. 51:562-567.

Day, P.R. 1986. Particle fractionation and particle size analysis. p. 545-567.
In A. Klute (ed.) Methods of soil analysis. Part 1. 2nd ed. Agron.
Monogr. 9. ASA, Madison, WI.

Grisso, R.D., C.E. Johnson, and A.C. Bailey, 1987. Soil compaction by
continuous deviatoric stress. Trans. ASAE. 30:1293-1301.

Gupta, S.C., and R.R. Allmaras. 1987. Models to access the susceptibility of
soil to excessive compaction. Adv. Soil Sci. 6:65-100.

Gupta, S.C., A. Hadas, and R.L. Schafer. 1989. Modeling soil mechanical
behavior during compaction. p. 137-152. In W.B. Larson, G.R. Blake,
R.R. Allmaras, W.B. Voorhees and S.C. Gupta (eds.). Mechanics and
related process in structured agricultural soils. Kluwer Academic
Publishers, The Netherlands.

Gupta, S.C., A. Hadas, W.B. Voorhees, D. Wolf, W.B. Larson, and EC.
Schneider. 1985 . Development of quids for estimating the ease of
compaction of world soils. Research Report, Binational Agric. Res.
Development, Bet Dagan, Israel. University of Minnesota. USA.

Gupta, S.C., E.C. Schneider, W.B. Larson, and A. Hadas. 1987. Influence of
corn residue on compression and compaction behavior of soils. Soil
Sci. Soc. Am. J. 51:207-212.

Harris, W.L. 1971. The soil compaction process. p. 9-44 In K.K. Barnes,
W.M. Carleton, H.M. Taylor, R.L. Throckmorton, and GE. Vanden
Berger. (eds.). Compaction of Agricultural Soils. Am. Soc. Agric.
Eng., St. Joseph, MI.

Hill, R.L., and M. Meza—Montalvo. 1990. Long- tem wheel traffic effects on
soil physical properties under different tillage systems. Soil Sci. Soc.
Am. J. 54:865-870.

Holtz, R.D., and W.D. Kovacs. 1981. An introduction to geotechnical
engineering. Prentice-Hall, Inc., Englewood Cliffs, NJ.

Horn, R. 1988. Compressibility of arable land. Catena Sup. 11:53-71.

44

Johnson, B.S., A.E. Erickson, and W.B. Voorhees. 1989. Physical conditions
of lake plain soil as affected by tillage and wheel traffic. Soil Sci. Soc.
Am. J. 53:1545-1551.

Kassa, Z. 1992. Pare water pressure and some associated mechanical
responses to uniaxial stress in structured agricultural soil. MS thesis.
University of Minnesota.

Klute, A. 1986. Water retention: laboratory methods. p. 635-662. In A. Klute
(ed.) Methods of soil analysis. Part 1. 2 nd ed. Agron. Monog. 9.
ASA, Madison, WI.

Larson, W.E., and S.C. Gupta. 1980. Estimating critical stress in unsaturated
soils from changes in pore water pressure during confined compression.
Soil Sci. Soc. Am. J. 44:1127-1132.

Larson, W.E., S.C. Gupta, and J.L.B. Culley. 1988. Changes in bulk density
and pore water pressure during soil compaction. Catena Sup. 11:123-
128.

Larson, W.E., S.C. Gupta, and R.A. Useche. 1980. Compression of
agricultural soils from eight soil orders. Soil Sci. Am. J. 44:450-457.

Martinson, O. K. 1993. N dynamics in and under a fixing dry bean using "N
as tracer. MS. Thesis, Michigan State University.

McNabb, D.H., and L. Boersma. 1993. Evaluation of the relationship between
compressibility and shear strength of Andisols. Soil Sci. Am. J.
57:923-929.

Meek, B.D., E.R. Rachel, L.M. Carter, and W.R. DeTar. 1992. Bulk density
of a sandy loam: traffic, tillage, and irrigation-methods effects. Soil
Sci. Soc. Am. J. 56:562-565.

Muller, L., P. Tille, and H. Krestschmer. 1990. Trafficability and workability
of alluvial clay soils in response to drainage. Soil Tillage Res. 16:273-
287.

O’Sullivan, M.F. 1992. Uniaxial compaction effects on soil physical properties
in relation to soil type and cultivation. Soil Tillage Res. 24:275-286.

Pidgeon, JD. and B.D. Soane. 1977. Effect of tillage and direct drilling on
' soil properties during the growing season in a lang-tem barley mono-
culture system. J. Agric. Sci. 88:431-442.

45

Pierce, F.J., M.C. Fortin, and M.J. Staton. 1992. Immediate and residual
effects of zone-tillage in rotation with no-tillage on soil physical
properties and corn perfomance. Soil Tillage Res. 24: 149-165.

Reinert, DJ. 1990. Soil structural fom and stability induced by tillage in a
Typic Hapludalf. Ph.D diss. Michigan State Univ. East Lansing.

Snedecor, G.W. and Cochran. 1967. Statistical methods. Sixth edition. The
Iowa State University Press, Ames, Iowa.

Soane, B.D., J.W. DicKson, and DJ. Campbell. 1982. Compaction by
agricultural vehicles: a review HI. Incidence and control of compaction
in crop production. Soil Tillage Res. 2:3-36.

Sowers, GP. 1986. Consistency. p. 391-399. In A. Klute (ed.) Methods of
soil analysis. Part 1. 2nd ed. Agron. Monogr. 9. ASA, Madison, WI.

Voorhees, W.B., C.G. Senst, and W.W. Nelson. 1978. Compaction and soil
structure modification by wheel traffic in the N orthem Corn Belt. Soil
Sci. Soc. Am. J. 42:344-349.

Xu, Chuanguo. 1994. ConservatiOn tillage, chemical input, and manure history
in regulating corn (Zea mays L.) and soybean (Glycine max (L). Merr.)
production and fate of nitrogen in soil. Ph. D. diss. , Michigan State
University.

46

Table 1. Soil properties of the Capac loam, Kalamazoo loam and Misteguay silty
clay soils averaged across treatments.

 

 

 

 

Soil LL‘i' PL PI OC N Clay Silt Sand GMD

(kg kg") (2 kg“) (mm)
0-3cm

Capac 0.25 0.17 0.08 17 1.6 110 340 550 0.076

Kalamazoo 0.22 0.15 0.07 11 1.1 90 350 560 0.032

Misteguay 0.53 0.26 0.27 31 2. 1 480 470 50 ----

n1 4 12 4 20 20 12 12 12 3

15-18cm

Capac 0.25 0.17 0.08 17 1.5 100 350 550 0.076

Kalamazoo 0.21 0.13 0.08 7 0.8 100 350 550 0.031

Misteguay 0.53 0.26 0.27 30 1.9 490 460 50 ---

n 4 12 4 19 19 12 12 12 3

 

1'LL = Liquid limit, PL = Plastic limit, P1 = Plasticity index, CC = Organic

carbon, N = Nitrogen, GMD = geometric mean diameter of sand particles.
in for LL and PI consisted of 1 measure for each treatment while 3 replications were

measured for each treatment for the other parameters.

47

Table 2. Bulk density prior to compression test for the Capac loam, and

Kalamazoo loam soils.

 

 

 

Soil Bulk Density (Mg m“)
NTTT CIT NTBT CTBT
0-3cm
Capac 1471:0021: l.57:i:0.02 1.38:1:001 1.37:0.02
Kalamazoo 1.55:0.03 1.50:0.03 1.491002 1.471003
15 - 18 cm
Capac 1.53:001 1.571002 1.501002 1.481001
Kalamazoo l.64:|:0.03 1.641001 l.63:i:0.02 l.66;t0.02

 

T NTT = No tillage in the track, CTT = Conventional tillage in the track,
NTBT = No tillage between tracks, CTBT = Conventional tillage between tracks.
2]: mean :1; standard error of the mean. Each value represents an average of 12

measurements.

48

 

 

Table 3. Coefficients of the regression of bulk density 0).) on soil
water content (0.) prior to compression test for the Misteguay
silty clay using the regression model (p, = a + b 0.).
Tillage/traffic Intercept (a) slope (b) R2
0-3em
NTT? 1.77 a - 1.29 ab 0.60
CTT 1.90 a - 1.68 a 0.84
NTBT 1.69 b - 1.35 ab 0.66
CTBT 1.47 c - 0.85 b 0.66
15 - 18 cm
NT 2.10 at - 2.09 a 0.96
CTT 1.96 ab - 1.64 a 0.82
NTBT 2.01 b - 1.94 a . 0.97
CTBT 1.98 b - 1.78 a 0.94

 

Coefficients followed by the same letter are not significantly different at

p = 0.05.

T NTT = No tillage in the track, CTT = Conventional tillage in the track,
NTBT = No tillage between tracks, CTBT = Conventional tillage between

tracks.

1 The NTT was not significantly different from CTT due to higher variation
at CTT (R2 = 0.82).
n = 16 for each regression.

49

Table 4. Comparison of regression equations of the fom (m = a + b0. + c0.2 )
for compression index (m) and soil water content (0.) for Capac loam,

 

 

and Misteguay silty clay.
Depth Tillagel'l‘raffic a b c R2 0... m... 11
(cm) flicks“) (Msnr’)
Capac loam
0-3 TT 0.08 2.63 - 7.78 0.49 0.17 0.30 24
BT 0.26 1.75 - 7.03 0.32 0.13 0.37 24
15-18 All 0.08 3.61 -12.59 0.45 0.14 0.34 48

Misteguay silty clay

0-3 T -0.16 3.53 - 7.46 0.28 0.24 0.26 24
BT 0.14 1.98 - 4.17 0.26 0.24 0.38 24
15-18 All 0.02 1.04 - 1.26 0.61 0.41 0.23 48

 

1' T = Track, BT = Between tracks, All = Track and between tracks combined
together.

50

Table 5. Comparison of regression equations of the fom (0, = 10“ ” " m) for
preconsolidation pressure (0,) and soil water content (0.) for Capac

loam, Kalamazoo loam, and Misteguay silty clay.

 

 

Tillage/Traffic Intercept (a) Slope (b) R2 Intercept (a) Slope (b) R2
Capac loam
0 - 3 cm 15 - 18 cm
NTT? 2.90 d - 3.23 c 0.92 2.97 e - 3.30 c 0.95
CIT 2.92 c - 4.11 d 0.98 3.01 d - 4.13 d 0.98
NTBT 2.87 c - 3.96 cd 0.94 3.07 cd - 4.86 de 0.95
CTBT 2.80 e - 4.30 d 0.94 3.17 c - 6.08 e 0.95
Kalamazoo loam
NTT 2.94 d - 4.96 c 0.93 3.07 d - 5.59 c 0.95
CTT 2.96 c - 7.36 d 0.89 3.12 c - 7.04 cd 0.88
NTBT 2.76 c - 5.06 c 0.89 3.05 c - 6.29 c 0.97
CTBT 2.90 c e - 6.94 d 0.93 3.15 c - 7.86 d 0.95
Misteguay silty clay
NTT 3.15 d - 3.56 d 0.91 2.97 d - 1.82 c 0.88
CTT 2.95 d - 1.86 c 0.90 2.90 c - 1.77 c 0.96
NTBT 3.32 c - 4.86 e 0.93 3.11 c e - 2.84 de 0.83
CTBT 3.04 d - 2.97 d 0.91 2.91 e - 2.08 c e 0.98

 

Coefficients followed by the same letter are not significantly different at p = 0.05.

1‘ NTT = No tillage in the track, CTT = Conventional tillage in the track, NTBT =
Na tillage between tracks, CTBT = Conventional tillage between tracks.

11 = 16 for each regression.

1000

100

10

\ilm (kPa)

1000

100

10

Figure 1.

51

 

 

 

 

 

 

and Misteguay silty clay soils at 0—3 cm and 15-18 cm depths for the
NTBT treatment.

' . l-~I Capac
- U I H Kalamazoo
I ‘ A
I \ " ' ' ‘ Misteguay
I \ I
1 N ‘
E \
I \
- \
-. \ .
\- ‘A 0 - 3 cm
'5 2‘-
‘. 15 - 18 cm
I l l
01 02 03 04 05
-1
9... (k9 kg )
Soil water characteristic curves for the Capac loam, Kalamazoo loam,

52

 

 

 

 

 

 

 

 

a Capac
6‘ 1.4 - =—-
E ‘l
g 1.6 - 9,, (kg kg’)
‘3 o 0.24
Q 1.3 - I 0.19
A 0.12
O 0.05
b
A 1.4 -
‘1’
E
Bi 1.6 -
é
a
0- 1.8 -
c
1.0 - _ A: =
'5 -1 ‘\
o. 1.1- 9,..(k9k9) \
\. 0 0.35
Q 1.2 - I 0.29
A 0.19
O 0.11
I I j 1 1 I U U' I I U T Tij'
10 100 1000
0' (kPa)
Figure 2. Soil compression curves for the Capac loam, and Misteguay silty clay

soils and nomalized compression curves for the Misteguay soils as
affected by water content (0.). (Error bars represent the standard error
of the mean; error bars for some points are masked by symbols due to
very small std error values). .

 

0.50 -

0.40 -

0.30 -

0.20 -

0.10 -

Capac
PL LL
E]
D D
E] o ‘3

 

 

a
. Track 0-3 cm

 

m (Mg m")

0.50 -

0.40 -

0.30 -

0.20 -

0.10 -

0.0
. D
O
' a.
O

 

15-18 cm

 

 

 

 

 

l I I I

0.00 0.05 0.10 0.15 0.20 0.25

Figure 3.

9. (k9 kg")

The relationship between the compression index (rn) and soil water
content (0.) for Capac loam for the 0-3 cm and 15-18 cm depths.

0.30

54

 

 

 

 

 

 

 

 

 

0 70 Kalamazoo
0 60 _ + m 0,, = 0.38 -0.810m R’= 0.37
”E 0.50 _ ‘9— m 1.-., = 0.23 - 0.31 0,,, R2 = 0.12
g 0'40 ' . ‘0. ‘ PL LL
v 0.30 -
E 0.20 -
0.10 -
0.0 0.3
-1
9.. (kg kg )
Misteguay
0.60 - 0 PL
I LL
”A 0.50 " U
E 0.40 - ‘
or 5:13.
2 0.30 - \\
V D
E 0.20 - ‘II! \m\c
0.10 - Track
0 Between track
I I I I l

 

 

 

 

0.0 0.1 0.2 0.3 0.4 0.5 0.6
9... (kg kg")

Figure 4. The relationship between the compression index (m) and soil water
content (0.) for the Kalamazoo loam and Misteguay silty clay for the 0-
3 cm depth.

55

 

 

 

 

\II... = - 6 kPa
1.2 -
1.4 -
1.6 -
1.8 -
A I Capac
n
'E 2.0 - 0 Kalamazoo
g A Misteguay
of III... = -100 kPa
1.2 -
1.4 -
"~
1.6 " \\
1.8 - \ \
2.0 -
I I I I I I I II I I I —I I I I I'
10 100 1000
0' (kPa)
Figure 5. Compression curves at - 6 kPa and - 100 kPa matric potential (41.) for

Capac loam, Kalamazoo loam, and Misteguay silty clay soils under
CTBT treatments at 0-3 cm depth. (Error bars represent the standard
error of the mean; error bars for some points are masked by symbols
due to small std error).

 

 

 

D—-- crer

 

1J2-
1A!-
115-
113-
21)-

pb (M9 "1.3)

Kalamazoo

 

1.2 -
1.4 -
1.6 -
1.8 a
2.0 -

 

 

 

I j I Ifiij' 1 I I IITIT'

10 100 1000

FigureG.

0' (kPa)

Compression curves at - 100 kPa matric potential 01'.) for Capac loam,
Kalamazoo loam, and Misteguay silty clay soils under different tillage
treatments at 0-3 cm depth. (Error bars represent the standard error of
the mean; error bars for some points are masked by symbols due to
smdlfldmmn)

 

 

 

 

 

 

 

 

600 -
0— NTT
400 ' o__ NTBT
200 - ""' cr'r
D—-- CTBT
Kalamazoo
A 600 -
CB .0
0L
£ 400 - & PL
b“ :
200 -
Misteguay
600 -
400 -
200 -
\U‘B-J
I
0.0 0.1 0.2 0.3 0.4
-1
9... (Re Re )
Figure 7. The relationship between the preconsolidation pressure (0,) and soil

water content (0.) for the Capac loam, Kalamazoo loam, and
Misteguay silty clay soils for the 0-3 cm depth for different tillage and

traffic positions.

 

0.5

58

 

+ 05,... =10"-"‘"-°'°-’ R’=0.90
800 - -o- usm.=10"-”"°”°-’ R’=0.94
---- uscTT =10"-°‘""’°-’ R’=0.82

—cr- uscm =10‘“°'°°”°-’ R2 = 0.79

 

 

 

 

 

9... (k9 k9")

Figure 8. Unconfined strength (US) as a function of soil water content (0.) for
the Capac loam for the 0-3 cm depth for different tillage and traffic

positions.

CHAPTERZ

ACCOUNTING FOR STRESS HISTORY IN MODELING SOIL COMPACTION

ABSTRACT

While much is known about the soil compaction process, current models do
not predict soil compressibility since they do not account for stress history and are not
linked to the field measurements. This study proposes a model of soil
compressibility, consisting of a stress history submodel that describes elastic,
recoverable deformation combined with a classical virgin compression submodel that
describes the plastic, unrecoverable deformation. The stress history model relates
preconsolidation pressure (0,) as a function of soil water content (0.) as
a, = 10" + an). The virgin compression model takes the form
pm = p, + m log (ah/a), where p. is bulk density and a is applied stress, and m
is the compression index modeled as a function of 0., as m = c + d0. + e03. The
stress history model predicted both a, and log“, a, reasonably well (R2 = 0.84 and
0.86) and (R2 = 0.78 and 0.89), respectively, for the data reported in the literature,
where ore is the critical stress. Field unconfined stress (US) measurements followed
the stress history model and were linearly related to a, (R2 > 0.98). A procedure was

proposed to construct field soil compression curves using field measurements of US,

59

6 0
p, and 0. in conjunction 0,, m, and are determined from laboratory measured soil
compression curves. It was also shown that a, is a good predictor of reported critical
strengths at which root elongation ceases. This study quantifies the importance of
stress history in modeling soil compaction and has immediate application in estimating
soil workability or traffieability.

61

INTRODUCTION

A critical concern with soil compaction is the determination of when the soil
istoowettotillortrafficandwhatdamagewilloccurtosoilwhenapplicdstresses
exceed the carrying eapacity of the soil. The soil compression curve is the basis for
such an understanding. While much is known about the compaction process (Barnes
et al., 1971; Gupta and Allmaras, 1987; Gupta et al., 1989), there are no studies that
have quantified the effect of drying on soil compressibility (McNabb and Boersma,
1993), particularly under field conditions. A soil based emphasis on modeling soil
compaction is the virgin compression curve which, by definition, defines plastic,
unrecoverable deformation, and is generally well described (Larson and Gupta, 1980;
Gupta et al., 1985; Horn, 1989). However, a soil is too wet and lor the stress
excessive if plastic deformation occurs. It is the region of elastic, recoverable
deformation (the secondary compression curve) within which a soil can be tilled or
trafficked without serious damage. It is this component of the soil compression curve
that reflects the stress history of soil and it is neglected in agriculture. By ’stress
history’ we mean that a soil has preserved, within its structure, remnants of previous
stresses and other changes it has experienced in the past that give it the ability to
sustain some level of stress without structural breakdown. Thus, a model that
predicts the maximum stress that a soil can withstand over a range of water contents
without causing soil compaction is very useful. Such a model will provide
information to whether a soil can be tilled or trafficked without soil damage.

In general, five different approaches have been used as the basis for modeling

62

the compression behavior of a soil: (1) the virgin compression curve (Soehne, 1958;
Bailey and VandenBerg, 1968; Bowen, 1975; Larson et al.,l980; Lebert and Horn,
1991; Bingner and Well, 1992), (2) the critieal stress (Larson and Gupta, 1980; Gupta
and Larson, 1982; Gupta et al. 1985); (3) the relationship between strain and applied
stress during triaxial tests (Bailey et al., 1984; Bailey et al., 1985; Bailey et al.,
1986; Grisso et al.,l987; Bailey and Johnson, 1989); (4) a finite element analysis
(Perumpral et al, 1971; Coleman and Perumpral, 1974; Pollock, Jr. et a1. 1986;
Gassman et al., 1989; Raper and Erbach, 1990 a; Raper and Erbach, 1990 b); and (5)
generalized curve fitting techniques (Blackwell and Soane, 1981; Howard et a1; 1981;
Leeson and Campbell, 1983; Angers et a1, 1987, Lebert et al., 1989; Canarache,
1991; Lebert and Horn, 1991). None of these models account for the stress history
of the soil, although Lebert et a1. (1989) and Lebert and Horn (1991) predict the
preconsolidation pressure (0,) from soil properties. The neglect of stress history in
current models may be related to the fact that compression tests are usually performed
on disturbed soil samples and at relatively high soil water contents, both of which
tend to mask the stress history of a soil.

The a, is an indication of the maximum previously applied stress sustained by
a soil (Holtz and Kovacs, 1981) and defines the limit of elastic deformation in the soil
compression curve. Thus, in agriculture, application of stress greater than the highest
previously applied stress should be avoided (Gupta et al. , 1989; Lebert and Horn,
1991) in order to avoid unrecoverable soil deformations. Since a, should be the

maximum stress applied to a soil to prevent further soil compaction, a model of a,

63

can form the basis of a stress history model.

This study proposes a two component model of soil compressibility, consisting
of a stress history submodel that describes elastic, recoverable deformation in terms
of 0,, and a virgin compression submodel, a submodel which describes the plastic,
non-recoverable deformation in terms of bulk density 00..) and applied stress (0‘); both
submodels are a function of soil water content (0.). Field unconfined stress (US)
measurements are related to a, and used in conjunction with the compression index
(m) and published values of critical stress (0,) to develop field based soil compression

curves .

MATERIAL AND METHODS
Model Development

A common basis for compaction models is the soil compression curve,
frequently expressed in terms of pg as a function of log a (Figure l). The general
position of this curve varies with soil type and 0., (Larson and Gupta, 1980; Larson et
al., 1980; Gupta et al., 1985; Gupta et al., 1987; Gupta and Allmaras, 1987; Lebert
and Horn, 1991). For agricultural soils that have experienced previous stress, the
compression curve consists of two distinct regions:/the secondary compression curve,
a region of small, elastic and recoverable deformation that defines the stress history of
a soil; and the virgin compression curve, a region of plastic and unrecoverable
deformations (Gupta et al, 1989; Lebert and Horn, 1991). The a, divides the

compression curve into these two regions (Lebert and Horn, 1991) and the slope of

64
the virgin compression curve is called the compression index (m) (Bradford and
Gupta, 1986).

The soil compaction model proposed herein estimates a soil compression curve
in terms of a stress history model and a virgin compression model (Figure 2). The
stress history model takes the general form of the relationship between a, and 0,.
(Figure 2a) expressed as
a, = 10" + m [1]
where a and b are fitted parameters. The regressions of log“, 0', on 0. (Equation [1])
varied by tillage and traffic treatment as reported in chapter 1. The coefficient of
determination (R’) of the regressions ranged from 0.83 to 0.98, the intercepts ranged
from 2.76 to 3.32, and the slopes ranged from -l.77 to -7.86. The virgin
compression model takes the general form

pa... = p. + m 108 (Om/0) [2]
where a is the applied stress (kPa) and m is the compression index (Figure 2b).
Although the virgin compression curves for a given soil have been reported to be
parallel, at least at high 0., (Larson and Gupta, 1980; Larson et al., 1980; Saini et
al., 1984; Hakansson et al., 1988; O’Sullivan, 1992), we found the virgin
compression curves were not always parallel (Figure 3). This agrees with
Schmertmann (1955) who reported that the curves for saturated soils intersect within a
narrow range of void ratio. For a given soil type, m is described as a function of 9,,
expressed as

m=c+d0.,+e0,,,2 [3]

65

The m... was found to occur near the plastic limit. Although this relationship was
weak and not consistent for all soils, it recognizes the variability of m, a portion of
which is explained by 0.. The variability in m needs further quantifieation.

The compaction model then describes the compressive behavior of soil as a
function of p.” a, 0,, 0,,, and soil management practices providing parameters for
using Equations [1], [2], and [3]. The model works in the following manner. For
applied stress less than the 0,, deformation is elastic so that wheel traffic will cause
no additional compaction. For applied stress greater than the 0,, deformation is
plastic, compaction increases in proportion to the applied stress, and the rate of
deformation m is a maximum near the plastic limit. Thus, the degree to which an
applied stress causes elastic or plastic deformation is largely a function of stress

history, 0., and soil management for a given soil type.

Model validation

The stress history component of the proposed compaction model was evaluated
relative to data reported by Larson and Gupta (1980), Reinert (1990), and Kassa
(1992). Data on 0.,, 0,,and a, were obtained from those studies. These data were
then fit to the stress history portion of the proposed compaction model. Field
validation of the stress history model was accomplished by evaluating field measured
penetrometer measurements reported in chapter 1 for the Capac loam (Fine loamy,
mixed, mesic Aerie Ochraqualfs) against a, predicted from Equation [1] . Appropriate

regressions were performed in Sigma Plot 1.02 (Jandel Scientific, P.O. Box 7005,

66

San Rafael, CA).

RESULTS AND DISCUSSION
Model Validation

The stress history model (Equation [1]) expressed in Figure 4 was obtained
from the conventional tillage treatment at the 0-3 cm depth in the Capac loam (110 g
kg" clay), the Kalamazoo loam (Fine loamy, mixed, mesic Typic Hapludalfs) (90 g
kg" clay), and the Misteguay silty clay (Fine, mixed (calcareous), mesic Aerie
Haplaquepts) (480 g kg" clay). The information was obtained between tracks in the
field. Also shown in Figure 4 are data reported by others. The stress history model
for the Kalamazoo loam predicted a, reasonably well (R2 = 0.84) for data reported by
Reinert (1990) for the same soil (Figure 53). The stress history model for the Capac
loam predicted a, of the Ves clay loam (300 g kg" clay) and the Webster clay loam
(330 g kg" clay) reported by Kassa (1992) well, with an R2 of 0.86 and a close fit to
the 1:1 line, even though the range of soil water used for the compression tests was at
the high end only (Figure 5b). Thus, the stress history model predicts the elastic
deformation of a soil reasonably well.

Larson and Gupta (1980) proposed the use of critical stress (0,) to define the
maximum stress a soil can withstand without damaging aggregates. The 0,
corresponds to the minimum pore water pressure at which soil aggregate ruptures and
occurs at a, > 0,. care was not measured for the Michigan soils. However, we

analyzed data from Kassa (1992) and found a strong linear relationship between log",

67
a, and 0,, with R2 ranging from 0.86 to 1.00 (Figure 6) given as
a, = 10" * "’) [4]
wherefandgarefittcdparameters. Thestresshistorymodel fortheCapacpredicted
the log“, a, well for data from Larson and Gupta (1980) and Kassa (1992), with 1?.2 of
0.78 and 0.89, respectively, even though these soils had higher clay contents than the
Capac (Figure 7). Thus, a, and a, are closely related, both increasing with decreasing
0.,. Additionally, we found that the relationship between unconfined strength (US), as
measured in the field with a pocket penetrometer (chapter 1), and 0. follows the
stress history model (Figure 8a) as
US = 10"I * 'm’ [5]
where h and i are fitted parameters. Thus, field measures of US and 0. can be used
to estimate 0,,... (Figure 8b) from
0,,... = j + k(U S) [6]
where j and k are fitted parameters.

The importance of these findings are that estimates of field soil compression
curves can be constructed from easily measured soil properties: US, 0,., and )0,i in
Equations [1-6]. This is possible because, by definition, a, divides the compression
curve into two regions, the virgin compression curve is log-linear, and a, < 0,.
Therefore, the secondary compression curve can be constructed from a linear line
segment between p“ and 0,, and the virgin compression curve ean be constructed
using a, and both m and. a, (Figure 1).

We have shown the importance of a, and its relationship to a,= and field

68

measured US, but have not explored the relationship between a, and root penetration.
Gerard et al. (1982) reported that soil strength was a function of soil water content,
voids, and clay content and that the critieal strength (MPa) at which root elongation
ceased was solely a function of clay content ( 96) expressed as

critieal strength = 18.57 clay"M9 [7]
We calculated the critical strength predicted by Equation ['7] for the Capac loam,
Kalamazoo loam, and Misteguay silty clay (clay contents 110, 90, and 480 g kg",
respectively) and regressed the predicted critical strength on a, predicted from
Equation [1] for dry soil at a constant 0., of 0.10 kg kg" and at 0., corresponding to a
matric potential of -1.5 MPa as reported in chapter 1 (Figure 9). The regression was
linear and the relationship was strong (R2 = 0.99 and 0. 83 respectively). Therefore,
a, is also a good predictor of the critical strength at which root elongation ceases and
implies that soils with a considerable stress history are more likely to inhibit root

growth.

CONCLUSIONS

The proposed soil compaction model accounts for stress history in terms of a,
as a function of 0.. The stress history model was a good predictor of a, and or, from
the literature and is a good predictor of critical strength for root elongation. Because
a, was closely related to field measured US, it was possible to construct soil
compression curves from field measurements of US, p.” and 0. with knowledge of

laboratory measured soil compression curves from which values of 0,, m, and

69
possibly a, ean be obtained. The importance of stress history in modeling soil
compaction is clear. Stress history models have immediate application in estimating

soil workability or trafficability for a range of soils and soil management conditions.

LET OF REFERENCES

Angers, D.A., B.D. Kay, and P.H. Groenevelt. 1987. Compaction
characteristics of a soil cropped to corn and bromegrass. Soil Sci. Soc.
Am. J. 51:779-783.

Bailey, A.C., and C.E. Johnson. 1989. A soil compaction model for
cylindrical stress state. Trans. ASAE 32:822-825.

Bailey, A.C., C.E. Johnson and R.L. Schafer. 1984. Hydrostatic compaction
of agricultural soils. Trans. ASAE 27:952-995.

Bailey, A.C., C.E., Johnson, and R.L., Schafer, 1986. A model for
agricultural soil compaction. J. Agric. Eng. Res. 33:257-262.

Bailey, A.C., C.E. Johnson, R.L. Schafer, T.A. Nichols, and R.D. Grisso.
1985. A compaction model for agricultural soil. p. 2:203-212. In
Taction and transport as related to cropping systems. Proc. Int. Conf.
Soil Dynamics. Auburn, AL. June 17-19, 1985. Natl. Tillage Mach.
Lab. and Agric. Eng. Dept., Alabama Exper. Sta., Auburn, AL.

Bailey, A.C., and GE. VandenBerg. 1968. Yielding by compaction and shear
in unsaturated soils. Trans. ASAE 11:307-311,3l7.

Barnes, K.K, W.M. Carleton, H.M. Taylor, R.I. Throckmorton, and GE.
Vanden Berg. 1971. Compaction of agricultural soils. ASAE. Monogr.,
St. Joseph, MI.

Binger, R.L., and LG. Wells. 1992. Compact - a reclamation soil compaction
model part 1. model development. Trans. ASAE 35 :405-413.

Blackwell, P.S., and B.O. Soane. 1981. A method of predicting bulk density

changes in fields soils resulting from compaction by agricultural traffic.
J. Soil Sci. 32:51-65.

70

71

Bowen, H.D. 1975. Simulation of soil compaction under tractor-implement
traffic. ASAE paper No. 75-1569. Am. Soc. Agric. Eng. St. Joseph,
MI.

Bradford, J.M. and S.C. Gupta. 1986. Compressibility. p. 479-492. In A.
Klute (ed.) Methods of soil analysis. Part 1. 2nd ed. Agron. Monog. 9.
ASA, Madison, WI.

Canarache, A. 1991. Factors and indices regarding excessive compactness of
agricultural soils. Soil Tillage Res. 19:145-164.

Coleman, G.R., and J.V. Perumpral. 1974. The finite element analysis of soil
compaction. Trans. ASAE 17:856-860.

Gassman, P.W., D.C. Erbach, and S.W. Melvin. 1989. Analysis of track and
wheel soil compaction. Trans. ASAE 32:23-29.

Gerard, C. J., P. Sexton, and G. Shaw. 1982. Physical factors influencing soil
strength and root growth. Agron. J. 74:875-879.

Grisso, R.D., C.E. Johnson, and A.C. Bailey. 1987. Soil compaction by
continuous deviatoric stress. Trans. ASAE 30: 1293-1301.

Gupta, S.C., and R.R. Allmaras. 1987. Models to access the susceptibility of
soil to excessive compaction. Adv. Soil Sci. 6:65-100.

Gupta, S.C., A. Hadas, and R.L. Schafer. 1989. Modeling soil mechanical
behavior during compaction. p. 137-152. In W.B. Larson, G.R. Blake,
R.R. Allmaras, W. B. Voorhees and S.C. Gupta (eds.). Mechanics and
related process in structured agricultural soils. Kluwer Academic
Publishers, The Netherlands.

Gupta, S.C., A. Hadas, W.B. Voorhees, D. Wolf, W.B. Larson, and EC.
Schneider. 1985. Development of quids for estimating the ease of
compaction of world soils. Research Report, Binational Agric. Res.
Development, Bet Dagan, Israel. University of Minessota, USA.

Gupta, S.C., E.C. Schneider, W.B. Larson, and A. Hadas. 1987. Influence of
corn residue on compression and compaction behavior of soils. Soil
Sci. Soc. Am. J. 51:207-212.

72

Gupta, S.C., and W.B. Larson. 1982. Modeling soil mechanical behavior
during tillage. p. 151-178. In P.W. Unger, D.M. Van Dorcn,Jr., F.D.
Whisler, and EL. Skidmore (eds.). Predicting tillage effects on soil
physieal properties and process. Spec. Pub. 44. Am. Soc. Agron.
Madison, WI.

HAkansson, I., W.B. Voorhees, and H. Riley. 1988. Vehicle and wheel factors
influencing soil compaction and crop response in different traffic
regimes. Soil Tillage Res. 11:239-282.

Holtz, R.D., and W.D. Kovacs. 1981. An introduction to geotechnieal
engineering. Prentice-Hall, Inc., Englewood Cliffs, NJ.

Horn, R. 1989. Strength of structured soils to loading - a review of process on
macro and microscale; European aspects. p. 9-22. In W.B. Larson,
G.R. Blake, R.R. Allmaras, W. B. Voorhees and S.C. Gupta (eds.).
Mechanics and related process in structured agricultural soils. Kluwer
Academic Publishers, The Netherlands.

Howard, R.F., M.J. Singer, and G.A. Frantz. 1981. Effects of soil properties,
water content, and compactive effort on the compaction of selected
California forest and range soils. Soil Sci. Soc. Am. J. 45:231-236.

Kassa, Z. 1992. Pore water pressure and some associated mechanieal
responses to uniaxial stress in structured agricultural soil. MS Thesis.
University of Minnesota.

Larson, W.E., and S.C. Gupta. 1980. Estimating critical stress in unsaturated
soils from changes in pore water pressure during confined compression.
Soil Sci. Soc. Am. J. 44:1127-1132.

Larson, W.E., S.C. Gupta, and R.A. Useche. 1980. Compression of
agricultural soils from eight soil orders. Soil Sci. Soc. Am. J. 44:450-

457.

Lebert, M., N. Burger, and Horn, R. 1989. Effects of dynamic and static
loading on compaction of structured soils. p. 73-80. In W.E. Larson,
G.R. Blake, R.R. Allmaras, W. B. Voorhees, and S.C. Gupta (eds.).
Mechanics related process in structured agricultural soils. NATO
applied sciences, vol. 172. Kluwer Academic Publishers, The

Netherlands.

Lebert, M., and Horn, R. 1991. A method to predict the mechanical strength
of agricultural soils. Soil Tillage Res. 19:275-286.

73

Leeson, 1.1., and DJ. Campbell. 1983. The variation of soil critical state
parameters with water content and its relevance to the compaction of
two agricultural soils. J. Soil Sci. 34:33-44.

McNabb, D.H., and L. Boersma. 1993. Evaluation of the relationship between
compressibility and shear strength of Andisols. Soil Sci. Soc. Am. J.
57:923-929.

O’Sullivan, M.F. 1992. Uniaxial compaction effects on soil physical properties
in relation to soil type and cultivation. Soil Tillage Res. 24:275-286.

Perumpral, J.V., J.B. Lijedahl, and W.H. Perloff. 1971. The finite element
method for predicting stress distribution and soil deformation under a
tractive device. Trans. ASAE 14:1184-1188.

Pollock, Jr. D., J .V. Perumpral, and T. Kuppusamy. 1986. Finite element
analysis of multipass effects of vehicles on soil compaction. Trans.
ASAE 29:45-50.

Raper, R.L., and D.C. Erbach. 1990 a. Prediction of soil stresses using the
finite element method. Trans. of ASAE 33:725-730.

Raper, R.L., and D.C. Erbach. 1990 b. Effect of variable linear elastic
parameters on finite element prediction of soil compaction. Trans.
ASAE 33:731-736.

Reinert, DJ. 1990. Soil structural form and stability induced by tillage in a
Typic Hapludalf. Ph.D diss. Michigan State Univ., East Lansing.

Saini, G.R., and T.L. Chow. 1984. Compactibility indexes of some
agricultural soils of New Brunswick, Canada. Soil Sci. 137:33-38.

Schmertmann, I.H. 1955. The undisturbed consolidation behavior of clay.
Trans. ASCE 120:1201-1233.

Soehne, W.H. 1958. Fundamentals of pressure distribution and soil
compaction under tractors tires. J. Agric. Eng. 276-291.

74

 

Secondary Virgin
compression compression
curve 9 curve

3.:
I

       

Rebound curve
(elastic deformations)

\ (plastic
\ deformations)

<— Bulk Density —

—— Laboratory compression curve
— — Predicted field compression curve

 

 

 

—— Log Applied Stress —-—:--

Figure l. The secondary compression, rebound, and virgin compression
components of a typical soil compression curve illustrating the position
of the preconsolidation pressure (a, , the critieal stress (are), the
compression index (m), and the shift down and to the left of the curve
with increasing soil water content (0,). The dashed line represents a
field compression curve constructed from the proposed model.

75

Stress History Model

 

 

 

 

 

 

  
    

 

 

 

.
—— 0p = 10(2.87 -3300.) R2 . 0.94 a
0 1 l 1 l I
0.00 0.05 0.10 0.15 0.20 0.25 0.30
9". (k9 k9")
Virgin Compression Model
1.4 '-
6‘ .\
E 1.6 -
m
E 1-3 ‘ 0 p,,,,=1.01+0.2slogo also.”
Of I p"_,,80.76+0.35 log 0 R380.”
2-0 " A pun-045+ 0.42109 0 also.”
0 p,m-0.s1+0.2slog 0 R2: 0.80 b
.T... . . .m,
100 1000
0' (kPa)

Figure 2. The stress history model (a) expressing preconsolidation pressure (0,)
as a function of soil water content (0.,); and the virgin compression
model (b) expressing bulk density (p,) as a function of applied stress (a)
of the 0-3 cm depth for the Capac loam at four different 0..

76

 

 

 

.
1.4 - “ .
«PA . . . . o
E
m 1.6 -
E -1
1 G... (k9 k9 )
0' O 0.24
1.8 _ I 0.19
A 0.12
O 0.05
- - - - 0'
p
. 1......l . .1...aal
10 100 1000
0' (kPa)
Figure 3. Soil compression curves expressing bulk density 00,) as a function of

applied stress (or) for the 0-3 cm depth for the Capac loam at four
different 0.,. The dashed line represents the line of the regression of
preconsolidation pressure (0,) as a function of soil water content (0.).

 

 

 

 

Anny: ab

Flsure 4

77

 

El Ves 8. Webster clay loam (Kassa,1992)

 

 

 

300 a A Critical stress (Larson 8. Gupta, 1980)
O Kalamazoo loam (Relnert,1990)
A Critical stress (Kassa,1992)
50° " — -- Misteguay silty clay
A — Kalamazoo loam
10
o. —— Capac loam
3‘. 400 -
o.
b .
.\
200 1 A \
\k A A \\
t9 \\.
0 _ 0.0% Ah
I l l l
0.0 0.1 0.2 0.3 0.4
-1
9... (k9 k9 )
Figure 4 Preconsolidation pressure (0,) (from Kassa, 1992 and Reinert, 1990)

and critieal stress (0,) (from Kassa, 1992 and Larson and Gupta, 1980)
each as a function of soil water content (0.) compared with a, predicted
from the stress history models obtained from the 0-3 cm depth of the
Capac loam, Kalamazoo loam, and Misteguay silty clay for the
conventionally tilled treatment.

 

0.5

78

 

 

. OP= -4.00 '0' 0.95 O" R2 = 0.34 /

 

"/ Kassa (1992)
/ b

 

 

 

l I I

0 100 200 300

Predicted O'p (kPa)

Predicted and measured (Reinert, 1990) values of preconsolidation
pressure (0,) using the stress history model for the 0-3 cm depth of
Kalamazoo loam (5a) and the stress history model, for the 0-3 cm depth
of the Capac loam to compare with measurements of Kassa (1992) (5b).
The stress history models used were from the conventionally tilled
treatment.

79

 

 

 

 

 

‘4— Ves 0.25 cm 0, . 10‘°-”* ”2 “J R2 = 1.00
fo— Ves 25-40 cm 0. = 10““ ”2 “v’ R’ - 1.00
'°‘ " Webster 0.25 cm (36 = 10"-°'* °-°’°-’ R’ = 0.97
1000 '3_U" Webster 2540 cm 0. = 10“-“*°-°’°-’ R2 = 0.88
I
7; .
a.
£5
0
b
100 -:
1
7 Cl
[5 Kassa (1992)
10 I I I I
0 20 40 60 80 100

(1'p (kPa)

Figure 6. The relationship between critical stress (0;) and preconsolidation
pressure (0,) from Kassa (1992).

80

 

 

Kassa (1992)
—. - - 6c: 10 (0334’ 0.01 6') R2 = 0.89
Larson 8. Gupta (1980)
1000 ---O-- cc=1o(1.30¢6.100.) R2=0.78
.s .0
A 4 -'
a a
Q I - ‘
a s/
o _ . 1 '
b 100 I ..-" O
o/."'
.7" O
O/ '
I
10 u 1 u u
0 50 100 150 200 250
Predicted O'p (kPa)
Figure 7. The relationship between critical stress (0,) measured by Kassa (1992)

 

and Larson and Gupta (1980) and preconsolidation pressure (0,)
predicted using the stress history model derived from the 0-3 cm depth
of Capac loam when conventionally tilled.

 

81

 

800
—o— “Mum-mo.» rat-0.80
A 4—03m-1m'wu R280]!
g 600 - Q non-am-wa-fl-wu 1:1-0.92
x CI? -A-- amI1W'm’J R330.“
a. .
b 400 -
L
O
"’ 200 -
D
0

 

 

 

 

 

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35
-1
9... (Re Re )

 

 

 
 
    

 

 

 

A ----- 8,...Tr - - 14.88 + 0.48 (08) + 2.08 (US)2 - 8.89 (03?
(I a - . .
n- 800 _ 0,,," 141 41 +1 some) 2 3
a — - cm a 22.87 + 8.01 (US) - 8.49 (us) + 1.38 (US)
— — o = 48.88 + 1.78 (US)
6‘ 800 - ”m
/
'5 //
.3 400 - /
'U -/
e 200 - . //
o. /
0 r l
0 200 400 600

Predicted U S (kPa)

Figure 8. Unconfined strength (US) or predicted preconsolidation pressure (0,) as
related to soil water content (0.,) in (a) and a, as predicted US from (b)
using data from the 0 - 3 cm depth of the Capac loam in the no-till-
track (NTT) and conventionally-tilled-between—track (CTBT)
treatments.

82

 

q
I

.. = 7.82 - 9.09 o, R2 = 0.99
+ cs 15,, m = 12.5 - 43.7 o", R2 = 0.83

 

h
J

3

Predicted Critical Strength (MPa)
OI
l

2

‘lKalamazoo
6

-IMisteguay O

Capac O

 

 

 

0

Figure 9.

I I I I I

.0 0.1 0.2 0.3 0.4 0.5 0.6

Predicted O'p (MPa)

The relationship between critical strength at which root elongation
ceases, (as predicted from Gerard etal., 1982) and predicted
preconsolidation pressure (0,) for 0. = 0.10 kg kg" or #1., at -1.5 MPa
matric potential for the 0 - 3 cm depth of the Capac loam, Kalamazoo
loam, and Misteguay silty clay soils for conventionally tilled between
tracks.

CHAPTER3

A SPREADSIIEET PROCEDURE FOR ESTIMATING PRECONSOLIDATION

PRESSURE FROM SOIL COMPRESSION CURVES

ABSTRACT

Classical graphics and regression procedures have been used to estimate
preconsolidation pressure a, from soil compression curves, but none are easy to use
and they often involve subjective judgement. This paper briefly reviews 9 methods
used to estimate 0,, describes a spreadsheet procedure for its estimation from soil
compression curves, and evaluates the spreadsheet procedure with classieal methods
and published data. A spreadsheet was developed in Quattro Pro, Version 4.0, to
calculate a, from soil compression curves. Five different estimation methods were
programmed into the spreadsheet, for an applied stress sequence of 25, 50, 100, 200,
400, 800, and 1600 kPa. The a, was determined above for each method and
compared to the a, estimated using the graphical procedure of Casagrande (1936) for
288 soil compression curves from three soils in Michigan and from values reported in
the literature. Some methods fit the data best at low a, (high soil water content)
while others fit the data better at high a, (low soil water content). Therefore, a

combination of methods was found to fit the experimental data best. Methods 1 and 3

83

84

determine a, as the intersection of the line that passes through the first two points, or
the regression line fitted to four points, respectively, in the secondary compression
portion of the compression curve and the extension of the virgin compression line
determined from the points associated with applied stress of 800 and 1600 kPa. The
final spreadsheet procedure provides a fast and reliable estimation of a, and eliminates

subjective judgment associated with classical graphieal procedures.

8 5

INTRODUCTION

The compressive behavior of soil is expressed graphieally in the relationship
between the logarithm of applied stress and some parameter related to the packing
state of soil, most often void ratio or bulk density (Casagrande, 1936; Leonards,
1962; Holtz and Kovacs, 1981). When no previous stress has been applied, this
relationship is theoretieally linear and the applied stress results in an unrecoverable
deformation (Larson and Gupta, 1980; Larson et al., 1980; Culley and Larson, 1987;
Gupta and Allmaras, 1987; Lebert and Horn, 1991). However, when a soil has
experienced a previous stress, a change in the stress acting on a soil will result in
some deformation, which can either be relatively small and recoverable or
unrecoverable (Stone and Larson, 1980; Gupta et al.,l989; Lebert and Horn, 1991).
As a result, the packing parameter versus log applied stress curve is still log-linear,
but much flatter. The term preconsolidation pressure has been used to denote the
”break” in the consolidation curve (Holtz and Kovacs, 1981; Jamiolkowski et al.,
1985) between these two cases. Thus, the preconsolidation pressure divides the soil
compression curve into a region of small, elastic and recoverable deformation
(secondary compression curve) and a region of plastic and unrecoverable deformation
(virgin compression curve).

In saturated soils, the preconsolidation pressure is used in settlement theory to
estimate the load support eapacity of soil (Leonards, 1962; Holtz and Kovacs, 1981).
In agricultural soils, loads are applied to unsaturated soils. In theory, stress history is

important to the compressive behavior of unsaturated soils since additional soil

86

compaction occurs only when the load exceeds the preconsolidation pressure (Gupta et
al., 1989; Lebert and Horn, 1991). Although the emphasis in soil compaction studies
has been on the non-recoverable deformation that occurs with applied stresses within
the range of the virgin compression curve (Larson et al., 1980; Gupta et al., 1989;
Lebert and Horn, 1991; Binger and Wells, 1992), the importance of stress history is
recognized, particularly as it relates to conservation tillage systems (Culley and
Larson, 1987; Larson et al., 1988). However, its importance in predicting soil
compaction and trafficability is poorly understood (Horn, 1989; Lebert and Horn,
1991; Binger and Well, 1992; McNabb and Boersma, 1993).

Preconsolidation pressure has been measured as part of recent soil compaction
studies (Culley and Larson, 1987; Lebert et al., 1989; Lebert and Horn, 1991).
However, its determination is somewhat imprecise. The most common methods in
classical soil mechanics, such as Casagrande (1936) and Schmertmann (1955), are
graphical and developed for saturated soils. These methods have been applied to
unsaturated soils and the Casagrande method remains a standard for comparison to
other methods (Jose et al. , 1989). Additional methods have been used to estimate
preconsolidation pressure in unsaturated soils, primarily involving regression (Lebert
et al., 1989; Reinert, 1990; Lebert and Horn, 1991), but none are considered standard
techniques. In all cases, none of the methods currently available are easy to use and
often involve subjective judgement.

This paper briefly reviews methods used to estimate preconsolidation pressure,

describes a spreadsheet procedure for estimating preconsolidation pressure from

87

uniaxial compression tests for unsaturated soil conditions, and evaluates the

spreadsheet procedure with classieal methods and published results.

Review of Current Methods

The break in slope of a consolidation curve is not always sharp, and some
methodology must be chosen to assign a best estimate of the presumed break
(preconsolidation pressure). Thus, there is no agreed upon method of determining the
preconsolidation pressure. However, according to Leonards (1962), the earliest and
most widely used procedure to determine preconsolidation pressure is the Casagrande
(1936) procedure. The following discussion briefly describes nine procedures for
determining the preconsolidation pressure. The graphical methods are illustrated in
Figure 1.

The Casagrande (1936) method involves selecting the point of minimum radius
of curvature. This is accomplished by drawing horizontal and tangent lines at this
point and bisecting the angle between them, then extending the straight line portion of
the virgin compression curve until it intersects the bisector of the angle (Figure 1).
The pressure corresponding to this point of intersection is the estimate of the
preconsolidation pressure.

Burmister (1951) proposed a procedure in which the unloading-reloading stress
cycle defines the slope of a typical unloading curve and the form and size of the
characteristic triangle on a semi-logarithmic plotting of the curve (Figure 1). By

shifting the unloading curve upward and parallel to itself to a point where a

8 8
geometrically similar triangle of the same vertical intercept is found, the
preconsolidation pressure can be determined. The preconsolidation pressure is equal
to the position of the vertical leg.

Schmertmann (1955) suggested a procedure in which a horizontal line is drawn
parallel to the log of applied stress from the initial void ratio to the existing vertical
overburden pressure (Figure 1). A line parallel to the rebound-reload curve is drawn
through the vertieal overburden pressure, and the laboratory initial virgin compression
curve is extended until it intersects either the initial void-ratio or the rebound line.
The intersection point is defined as the preconsolidation pressure.

Sallfors (1975, as cited by Larson, 1986) used a method in which the two
straight parts of the stress-strain curve are extended and intersected (Figure 1). An
isosceles triangle is inscribed between the lines and the stress-strain curves. The
intersection point between the base of the triangle and the upper line represents the
preconsolidation pressure.

Anderson and Lukas (1981) predict the preconsolidation pressure (0,) from the

undrained shear strength (Su) and the effective vertical overburden pressure (p’):

a, = Su/(Su/p’) [l]

Culley and Larson (1987) used a statistical procedure to estimate the

preconsolidation pressure. First, a least square regression was determined considering

that all points lay on the virgin compression curve. Next, the compression curve was

8 9
divided in two regions assuming an initial estimate of preconsolidation pressure of 15
kPa. Regression equations for each region were then developed and a combined sums
of square calculated. The estimated preconsolidation pressure was then incrementally
increased by 5 kPa and the statistics recalculated. The procedure was repeated until
the lowest residual sums of squares was achieved.

Jose et al. (1989) used a log-log method in which the applied pressure and
corresponding void ratio are plotted in logarithmic scale for each segment of the curve
(Figure l). The preconsolidation pressure is assumed to be equal to the applied
pressure at the intersection of these two distinct lines. The authors did not reveal
their criteria for choosing which points were included in the calculation of the two
lines.

Lebert and Horn (1991) estimated the preconsolidation pressure as the
intersection of the regression lines fitted through the secondary compression curve and
the virgin compression curve (Figure 1). The authors did not reveal their criteria for

choosing which points were included in the calculation of the two lines.

MATERIAL AND METHODS
Spreadsheet Procedure

A spreadsheet was developed in Quattro Pro (Version 4.0, Borland
International, Inc. , Scotts Valley, CA, USA) to calculate the preconsolidation pressure
from soil compression curves. Equivalent procedures could be programmed in other

modern spreadsheets. Five different estimation methods were programmed into the

90

spreadsheet, for an applied stress sequence of 25, 50, 100, 200, 400, 800, and 1600
kPa. The first four methods estimated the preconsolidation pressure as the
intersection of two lines: (a) one that passes through the first two points, or the
regression line fitted to three, four, or five points, respectively, in the secondary
compression portion of the compression curve and (b) the extension of the virgin
compression line determined from the points associated with applied stress of 800 and
1600 kPa (Figure 2). Method 5 consisted of the Schmertmann (1955) method (Figure
1). The user simply enters the values of bulk density for the corresponding applied
stress and the regressions are performed by entering the advanced math/regression
menu under the tool subheading in Quattro Pro and executing the regression function.
The preconsolidation pressure was determined above for each method and
compared to the preconsolidation pressure estimated using the graphical procedure of
Casagrande (1936) for our data or from the preconsolidation pressure reported in the
literature for selected studies. Our data included 288 compression curves determined
as part of a study to evaluate the effects of tillage and wheel traffic on the
compressive behavior of three soils in Michigan. The soil samples used are from
experimental research plots managed under long term no—tillage and plowed plots
including the Kalamazoo loam (Fine loamy, mixed, mesic, Typic Hapludalfs ) loeatcd
at Kalamazoo, MI, the Capac loam (Fine loamy, mixed, mesic, Aeric Ochraqualfs)
loeated at East Lansing, MI, and the Misteguay silty clay (Fine, mixed (calcareous),
mesic, Aerie Haplaquepts) located at Saginaw, MI. These soils had been cropped in

no-tillage management for the last 13, 14, and 9 years, respectively. Measurements

91

from the literature were taken from studies by Burmister (1951), Crawford (1964),
Jose et al. (1989), Reinert (1990), and Kassa (1992). The relationships between
applied stress and deformation were obtained by carefully extracting data from the
graphics in those references. The methods were evaluated based on regression of 0,,
deternrined with the Casagrande method, on 8,, determined by a given method, and
neamess of the regression line to the 1:1 line. Based on these regressions, a single

spreadsheet procedure was developed for unsaturated soil conditions.

RESULTS AND DISCUSSION

The graphical construction suggested by Casagrande (1936) is based in the
choice of the point in the consolidation curve with minimum radius of curvature.
Research has shown that as soil sample disturbance increases, the selection of this
point is increasingly more difficult and the preconsolidation pressure will be lower
than those obtained for undisturbed soil samples (Schmertmann, 1955; Brumund et
al. , 1976; Holtz and Kovacs, 1981). However, using undisturbed soil samples, the
selection of the point of minimum radius can also be difficult to determine at high
water content because the compression curve is almost linear (Figure 3). This could
result in an overestimation of the preconsolidation pressure when compared with the
values of minimum preconsolidation pressure determined according to Schmertmann
(1955 - method 5).

As water content changes, the shape of the compression curve changes so that

the number of points in the secondary or virgin compression portion of the curve

92

changes (Figure 3). Therefore, a spreadsheet procedure to estimate the
preconsolidation pressure should consider the possibility of changing the number of
points that belong to the secondary compression curve in the fitting of the regression
line. In addition, as the soil dries, the virgin compression curve is shifted up and to
the right in a such away that for the lower water contents, only two points remain in
the virgin compression curve for applied stress of 800 and 1600 kPa. Thus, if the
procedures used by Culley and Larson (1987), Jose et al. (1989) and Lebert and Horn
(1991) are used to estimate the preconsolidation pressure for a range of water
contents, the preconsolidation pressure will be underestimated.

The regressions of predicted versus Casagrande method determined
preconsolidation pressures for the 288 soil samples from the Michigan tillage studies
are given in Figure 4. We evaluated overall performance of each method by
examining the coefficient of determination (R2) of the regression and the neamess of
the regression line to the 1:1 line. Method 1 had the highest R2 of 0.87 but tended to
underpredict relative to the 1:1 line at preconsolidation pressures above 200 kPa (soil
matric potentials < -100 kPa). Method 5 (Schmertmann, 1955) had a similar R2 to
method 1 and appeared to predict well at low preconsolidation pressures. However,
all points were above the 1:1 line. Methods 2, 3, and 4 tended to over predict at low
preconsolidation pressures (high water content) but did a better job at predicting at
higher preconsolidation pressures (lower water contents). Since the performance of
the methods varied depending on the range of preconsolidation pressures (and,

therefore, water contents), the methods 1 and 5 were combined with methods 2 and 3

93
and the regression analyses ealculated. Methods 1 and 5 were used to ealculate

preconsolidation pressures for matric potentials > -100 kPa and methods 2 and 3
were used for matric potentials < -100 kPa. This matric potential was chosen
because it corresponded to one of the four potentials used in our compression
measurements and preconsolidation pressures in the > -100 kPa matric potentials
were generally below 200 kPa pressure. By inspection of Figure 4, methods 1 and 5
predicted well below 200 kPa and methods 2 and 3 predicted best above 200 kPa.
All combinations improved R2 to 0.90 to 0.92 (Figure 5). However, the combination
of methods 1 and 3 showed the best correspondence to the 1:1 line. Therefore, the
combination of method 1 and 3 was chosen as the best method for estimation of the
preconsolidation pressure for unsaturated soil conditions for use in the final
spreadsheet (Figure 6).

Table 1 shows the preconsolidation pressure obtained from the current
literature and those estimated using methods 1 through 5. The regressions were
performed for both saturated and unsaturated soil conditions, and for saturated and
unsaturated combined (Table 2). For saturated, unsaturated, and combined
regressions, all methods predicted the preconsolidation pressure well, but methods 1,
2, and 3 showed close correspondence to the 1:1 line, with slopes near 1 and
intercepts near 0. The small difference between preconsolidation pressure obtained by
methods 1, 2 and 3 and those from the literature was probably due to the well defined
break point in the reported consolidation or compression curves. Also, the soil water

contents evaluated in these studies were high and the range was narrow compared to

94

the water content range evaluated in our soils. Therefore, the best overall method

observed was the combination of method 1 and 3.

Spreadsheet Procedure Overview

The spreadsheet procedure is given in Appendix I. The spreadsheet screen is
reproduced in Figure 6 and the regression plot is illustrated in Figure 2. The first
step is to load the spreadsheet cell commands into the spreadsheet program in the
order presented in Appendix I. For example, cell Al is the heading for column A.
Cell 62 is the equation to calculate the slope of the secondary compression curve.
Once loaded, the spreadsheet will calculate all the necessary parameters for the
preconsolidation pressure. First, type the bulk density corresponding to the applied
pressures in the spreadsheet. The user enters ”Tools” and then ”Advanced Math"
than ”Regression" and enter ”Go" . This updates the spreadsheet for the regression
output, the preconsolidation pressure, and the corresponding bulk density. At the
same time, a graphic plot similar to the form in Figure 2 is redrawn and can be
viewed by the user in the Graphics subdirectory (”View"). The user can alter the

spreadsheet to different applied loads once the proposed spreadsheet has been entered.

CONCLUSIONS
For unsaturated soil conditions, the preconsolidation pressure can be estimated
by using a spreadsheet procedure which uses a combination of method 1 for moisture

conditions at matric potential higher than or equal to - 100 kPa, and method 3 for

95

moisture conditions at matric potential lower than - 100 kPa. Preconsolidation
pressures estimated with this procedure corresponded to standard graphieal methods
and literature values. This spreadsheet procedure, provides a fast and reliable
estimation of the preconsolidation pressure. In addition, when used in the analysis of
data for a research project involving 8,, the use of a consistent, repeatable procedure
rather than a graphical procedure will eliminate one source of variability, such as

subjective judgment associated with classical graphical procedures.

LIST OF REFERENCES

Anderson, T.C., and R.G. Lukas. 1981. Preconsolidation pressure predicted
using Su/p’ ratio. p. 502-515. In Yong, R.N. and EC. Townsend
(eds.) Laboratory shear strength of soil. Symposium ASTM. Special
Technical Publieation 740. Chicago, Ill, 25 June 1980. Philadelphia,
Pa.

Binger, R.L., and LG. Wells. 1992. Compact - a reclamation soil compaction
model part 1. model development. Trans. ASAE 35:405-413.

Borland International, Inc. Scotts Valley, CA, USA.

Brumund, W.F., E. Jonas, and C.C. Ladd. 1976. Estimating in situ maximum
past (preconsolidation) pressure of saturated clays from results of
laboratory consolidometer test. p. 4-12. In Transportation Research
Board, National Research Council. Estimation of consolidation
settlement. Special Report 163. National Academy of Sciences.
Washington, D.C.

Burmister, D. 1951. The application of controlled test methods in
consolidation testing. p. 83-98. In Fifty-Fourth Annual Meeting of the
ASTM. Symposium on consolidation testing of soils. Special Technical
Publication 126. Atlantic City, N .J . June 18, 1951. Philadelphia, Pa.

Casagrande, A. 1936. The determination of the pre-consolidation load and its
practical significance. p. 60-64. In Int. Conf. on Soil Mech. and
Found. Eng. Proc. of ICSMFE. Cambridge, Mass. June 22-26, 1936.
vol. 3. Cambridge, Mass.

Crawford, C.B. 1964. Interpretation of the consolidation test. p. 93-108. In
ASCE, Soil Mechanics and Foundation Division. Design of foundations
for control of settlement. Proc. of the ASCE, Evanston, ILL. June 16-
19, 1964. Evanston, ILL.

Culley, J .L.B., and WE Larson. 1987. Susceptibility to compression of a
clay loam Haplaquoll. Soil Sci. Soc. Am. J. 51:562-567.

96

97

Gupta, S.C., and R.R. Allmaras. 1987. Models to access the susceptibility of
soil to excessive compaction. Adv. Soil Sci. 6:65-100.

Gupta, S.C., A. Hadas, and R.L. Schafer. 1989. Modeling soil mechanical
behavior during compaction. p. 137-152. In Larson, W.E., G.R. Blake,
R.R. Allmaras, W.B. Voohees, and S.C. Gupta (eds.). Mcchanieal and
related process in structured agricultural soils. NATO applied sciences
172. Kluwer Academic Publishers, The Netherlands.

Holtz, R.D., and W.D. Kovacs. 1981. An introduction to geotechnical
engineering. Prentice-Hall, Inc. , Englewood Cliffs. NJ.

Horn, R., 1989. Strength of structured soils due to loading - A review of
process on macro and microscale; European aspects. p. 9-22. In W.B.
Larson, G.R. Blake, R.R. Allmaras, W.B. Voohees and S.C. Gupta
(eds.), Mechanical and related process in structured agricultural soils.
NATO applied sciences 172. Kluwer Academic Publishers, The
Netherlands.

Jamiolkowski, M., C.C. Ladd, J.T. Germaine, and R. Iancellotta. 1985. New
development in field and laboratory testing of soils. p. 57-153. In
Publications Committee of XI ICSMFE (ed.). Proc. of the Eleventh Int.
Conf. on Soil Mech. and Found. Eng. San Francisco, CA, 12-16
August 1985. Netherlands.

Jose, B.T., A. Sridharan, and B.M. Abraham. 1989. Log-log method for
determination of preconsolidation pressure. Geotechnical Testing
Journal. 12:230-237.

Kassa, Z. 1992. Pore water pressure and some associated mechanical
responses to uniaxial stress in structured agricultural soils. M.S. thesis.
University of Minnesota.

Larson, R. 1986. Consolidation of soft soils. Swedish Geotechnical Institute.
Report 29. Linkiiping, Swedish.

Larson, W.E., and S.C. Gupta. 1980. Estimating critical stress in unsaturated
soils from changes in pore water pressure during confined compression.
Soil Sci. Soc. Am. J. 44:1127-1132.

Larson, W.E., S.C. Gupta, and J .L.B. Culley. 1988. Changes in bulk density
and pore water pressure during soil compaction. Catena Sup. 11:123-
128.

98

Larson, W.E., S.C. Gupta, and R.A. Useche. 1980. Compression of
agricultural soils from eight soil orders. Soil Sci. Soc. Am.J. 44:450-
457.

Lebert, M., N. Burger, and R. Horn. 1989. Effects of dynamic and static
loading on compaction of structured soils. p. 73-80. In W.B. Larson,
G.R. Blake, R.R. Allmaras, W.B. Voohees and S.C. Gupta (eds.),
Mechanical and related process in structured agricultural soils. NATO
applied sciences 172. Kluwer Academic Publishers, The Netherlands.

Lebert, M., and R. Horn. 1991. A method to predict the mechanieal strength
of agricultural soils. Soil Tillage Res. 19:275-286.

Leonards, G.A. 1962. Foundation Engineering. McGraw Hill Book Company,
Inc., NY.

McNabb, D.H., and L. Boersma. 1993. Evaluation of the relationship between
compressibility and shear strength of Andisols. Soil Sci. Soc. Am. J.
57:923-929.

Reinert, DJ. 1990. Soil structural form and stability induced by tillage in a
Typic Hapludalf. Ph.D. diss. Michigan State Univ., East Lansing.

Sillfors, G. , 1975. Preconsolidation pressure of soft high plastic clays. Thesis.
Department of Geotechnical Engineering. Gothenburg.

Schmertmann, J .H. 1955. The undisturbed consolidation behavior of clay.
Trans. ASCE 120:1201-1233.

Stone, J .A., and W.B. Larson. 1980. Rebound of five one-dimensionally
compressed unsaturated granular soils. Soil Sci. Soc. Am. J. 44:819-
822.

99

Table l. Preconsolidation pressure (0,) obtained from current literature and
using method 1 through 5 for saturated and unsaturated soil conditions.

 

 

 

Reference Preconsolidation Pressure (kPa)
Literature 1 2 3 4 5
Saturated

Burmistcr, 1951 Burmister 75 81 89 109 155 71
Burmister 350 372 351 360 441 270
Crawford, 1964 Casagrande 300 278 291 311 358 271
Casagrande 262 238 256 289 343 224
Jose et al., 1989 Log - log 105 95 95 102 111 90
Log - log 114 103 99 99 105 92
Log - log 120 126 126 126 128 120
Log - log 102 98 101 111 120 92
Unsaturated
Reinert, 1990 Casagrande 174 172 168 163 183 100

Casagrande 134 139 117 138 178 89

Casagrande 61 68 59 81 116 37
Casagrande 17 14 13 1 1 7 11
Kassa, 1992 Statistical 94 95 94 104 138 29

Statistical 82 73 92 126 156 18
Statistical 63 60 63 69 79 31

Statistical 32 29 31 34 35 9

Table l (cont’d)

Statistical
Statistical
Statistical

Statistical

70

bi

20

100

63
37
23
18

67
45
25

21

79
5 1
29
26

118
71
43

38

16

10

 

101

 

 

Table 2. Regression equations of preconsolidation pressure (0,) from current
literature and as determined by methods 1 through 5 .
Method Regression equations R2
Saturated
l o,(Literature) = 7.92 + 0.98 0r,(method 1) 0.98
2 a,(Literaturc) = 0.66 + 1.01 0‘,(method 2) 0.99
3 a,(Literature) = -1.99 + 0.96 0,(mcthod 3) 0.98
4 0,,(Literature) = 10.01 + 0.77 a,(method 4) 0.95
5 a,(Literature) = -11.13 + 1.23 0,(method 5) 0.98
Unsaturated
1 a,(Literature) = 3.50 + 0.97 a,(method l) 0.99
2 0,,(Literature) = -1.81 + 1.05 c,(method 2) 0.98
3 a,(Literature) = -4.16 + 0.95 0‘,(method 3) 0.92
4 a,(Literature) = -4.00 + 0.74 a,(method 4) 0.85
5 a,(Literature) = 15.66 + 1.46 a,(method 5) 0.82
Saturated & Unsaturated
l a,(Literature) = 3.41 + 1.00 a,(method 1) 0.99
2 0,(Literature) = 0.12 + 1.02 0,,(method 2) 0.99
3 0,(Literature) = 4.78 + 0.97 a,(method 3) 0.98
4 0,,(Literature) = -5 .13 + 0.80 c,(method 4) 0.95
5 0,(L.Iterature) = 21.05 + 1.10 a,(method 5) 0.94

 

Void Ratio Void Ratio

Log Void Ratio

Figure 1.

 

 

Casagrande (1936)

 

 

Log Applied Stress (kPa)

 

 

 

 

Schmertmann (1955)

 

Log Applied Stress (kPa)

 

 

fitted lines

 

 

Jose et al., (1989)

 

Log Applied Stress (kPa)

Compression (°/o) Void Ratio

Void Ratio

 

 

 

 

Bannister (1 951)

 

Log Applied Stress (kPa)

 

 

 

 

Siillfors (1975)

 

Applied Stress (kPa)

 

 
     

fitted lines

 

 

Lebert and Horn (1991)

 

Log Applied Stress (kPa)

Illustration of published methods for determination of the
preconsolidation pressure (0,) for soil compression curves.

103

 

 

Method 1 Method 2

 

<—— Bulk Density

 

 

 

 

Method 3 Illlethod 4

 

 

— Applied Stress —->

Figure 2. Illustration of methods 1 through 4 for determination of the
preconsolidation pressure (0,) for soil compression curves.

104

 

“

 

 

 

6‘
e
51 1.6 -
E .1
1 9... (kg kg )
0' O 0.24
1.8 _ I 0.19
0.12
O 0.05
I I I I I I I1l I I I I I I I I'
10 100 1000

0' (kPa)

Figure 3. The effect of water content on the soil compression curves for a Capac
loam soil.

105

 

    
 
 
 

a”! - 64.41 + 1.49 op. .
1 .1
Capac
Misteguay
Kalamazoo

fi=0w

Method 1

  

one I - 125.57 + 1.49 op-

     
  

 

 

    

 

 

 

 

A
(U
D.
x
V
0
o.
b L.
ope = - 346.83 + 1.54 op.
600 -
400 '-
200 '-
0 Method 4
800 - o, = 84.31 + 1.38 o"
800- ° . .- a 1:1
R’=08s “no weir ,4“
400- I(;::7":‘ ' u
A .‘ 4 It“: . I
200 - ragg'u'fi.
~53" " Method 5
0 l 1 I I
0 200 400 600 800
on»! (kPa)
FiEllre 4. Regression of preconsolidation pressure determined by the Casagrande

(1936) procedure (ow) on preconsolidation pressure estimated by
methods 1 through 5 (0,“) for 288 compression curves from three soil
series in Michigan.

106

 

6pc (kPa)

 
     
   
  
    
   
   

op: = - 54.17 + 1.35 op.

I I
2 g"
9‘. l
R = o. I
I
- i

1 :1
D Capac

‘ Misteguay
' Kalamazoo

Method 1 8 2

 

Method 1 8 3

 

Mehtod 5 8: 2

 

.
fl
"

 

    

Method 5 8 3
l I l l

 

 

0 200 400 600 800

Figure 5.

0pm (kPa)

Regression of preconsolidation pressure determined by the Casagrande
(1936) procedure (aw) on preconsolidation pressure estimated by
combinations of methods 1 and 5 with methods 2 and 3 (0,“) for 288
compression curves from three soil series in Michigan.

107

 

 

 

 

LOAD LOG LOAD BULK DENS B.Dvcc B.D ** METHOD 1
(kPa) (Mgr-1) won") (Mgm' )
m, = 0.0309
25 1.3979 1.5531 1.2632 1.5462 x = 2.3698
50 1.6990 1.5624 1.3623 1.5681
100 2.0000 1.5809 1.4614 1.5900 0, = 234
200 2.3010 1.6199 1.5605 1.6119 )0, = 1.58
400 2.6021 1.6779 1.6596 1.6338
800 2.9031 1.7587 1.7587 1.6557
1600 3.2041 1.8578 1.8578 " METHOD 3
mm = 0.3292
Regression Output: B.D,cc x = 2.5015
(Mgnr’)
Constant 1.4446 g, = 317
Std Err of Y Est 0.0107 1.5531 p, = 1.63
R Squared 0.9135 1.5624
No. of Observations 4 1.5717
Degrees of Freedom 2 1.5810
1.5903
X Coefficient (8) 0.0727 1.5996
Std Err of Coef. 0.0158
Figure 6. Reproduction of the computer screen of the spreadsheet for

deternrination of the preconsolidation pressure for soil compression
curves.

SUMMARY AND CONCLUSIONS

This study assessed the effect of stress history on the compression behavior of
three Michigan soils in response to changes in soil properties induced by tillage and
wheel traffic; proposed a two component model of soil compressibility that accounts
for stress history, and presented a spreadsheet procedure for estimation of the
preconsolidation pressure.

Changes in soil properties induced by tillage and wheel traffic affected the
compressive behavior of these three soils. Soil moisture regulated the shape of the
compression curve, while initial bulk density regulated its position. The initial bulk
density of the Misteguay silty clay, and subsequently the compressive behavior, was
greatly affected by soil water content, and required a normalization of the
compression curves to fit the generalized relationship of shifts in soil compression
curves with changes in soil water content. In general, no-tillage shifted the
compression curves, increased a, in the Capac and Kalamazoo soils but not in the
Misteguay, and had little effect on m in any of the soils. No-tillage also
corresponded to higher field measured unconfined strength than CI‘ in the Capac soil.
Wheel traffic shifted the position of the compression curves, due to their influence on
initial conditions, increased 0,, and decreased m. These shifts would support the

notion of improved trafficability on no-tilled and trafficked soils. No-tillage had some

108

109

effect, but wheel traffic did more to decrease the susceptibility of these soils to further
compaction by decreasing m and increasing 8,. Specifically, wheel traffic in no-
tillage(NTI')hadahigher8,intheCapacandKalamazoosoil,althoughCTI‘was
higher in the Misteguay soil. The perception of increased traffieability of soils in no-
tillage, as reported by farmers, relates not so much to tillage-induced differences in
soil physical properties but is primarily due to wheel traffic effects and the fact that
controlled traffic is likely in long-term NT. The other source of improved
trafficability would be associated with improved drainage if this were the ease in soils
under long-term no-tillage management. Soils that dry faster would support higher
loads earlier. Therefore, farmers should not only consider the adoption of controlled
traffic patterns to reduce overall soil compaction, but should focus mainly on the
enhanced resistance due to decrease in water content.

The proposed soil compaction model accounts for stress history in terms of 8,
as a function of 0.. The stress history model predicted reasonable values of 8, and 8c
from the literature and was a good predictor of critical strength for root elongation.
Because 8, was closely related to field measured US, it was possible to construct soil
compression curves from field measurements of US, p,, and 0. with knowledge of
laboratory measured soil compression curves from which values of 8,, m, and
possibly 8,= can be obtained. This model has immediate application in estimating soil
workability or trafficability for a range of soils and soil management conditions using
currently available soil management models.

For unsaturated soil conditions, the preconsolidation pressure can be estimated

110

by using a spreadsheet procedure which uses a combination of method 1 for moisture
conditions at matric potential higher than or equal to - 100 kPa, and method 3 for
moisture conditions at matric potential lower than - 100 kPa. Preconsolidation
pressures estimated with this procedure corresponded to standard graphical methods
and literature values. This spreadsheet procedure, provide a fast and reliable
estimation of the preconsolidation pressure. In addition, when used in the analysis of
data for a research project involving 8,, the use of a consistent, repeatable procedure
rather than a graphical procedure will eliminate one source of variability, such as
subjective judgment associated with classical graphical procedures.

Future research should be conducted to link the model developed in this study
with currently available soil management models in order to generate trafficability /
workability maps using available computer mapping programs. These maps will be a

useful tool for farmers uses in order to avoid soil compaction.

APPENDIX 1: Cells of the suggested spreadsheet procedure for estimation of
the preconsolidation pressure from soil compression curves.

A1: ‘LOAD

Bl: ‘LOG LOAD

Cl: [W11] ‘BULK DENS

D1: [W9] “B.D retav

E1: [W9] “B.D reg

F1: [W16] "'1' METHOD 1

F2: [W16] 'Cec =

62: (F4) (C4-C3)/(B4—B3)

A3: 25

B3: (F4) @LOG(A3)

C3: (F4) [W11] 1.5531

D3: (F4) [W9] (GSlO*(B3-B$8)+C$8)

E3: (F4) [W9] (D$lZ+C$18*BB)

F3: [W16] 'x a

G3: (F4) (GZ'(-B4)+C4-C9-GlO*
(-B9))l(GlO-GZ)

A4: 50

: (F4) @LOG(A4)

: (F4) [W11] 1.5624

: (F4) [W9] (GSIO‘(B4-B$8)+C$8)

: (F4) [W9] (D$12+C$18*B4)

: 100

= (F4) 01-06(15)

: (F4) [W11] 1.5809

: (F4) [W9] (6810*(BS-BS8)+C$8)

: (F4) [W9] (DSl2-l-C3181'B5)

F5: [W16] ’Prec press reta=

GS: (F0) 10‘GS3

A6: 200

B6: (F4) @LOG(A6)

C6: (F4) [W11] 1.6199

D6: (F4) [W9] (GSlO*(B6-B$8)+C$8)

E6: (F4) [W9] (D312+C$l8*B6)

F6: [W16] ’Bulk Dens reta =

G6: (FZ) (GZ‘(@LOG(G$5)-B4) + C4)

A7: 400

B7: (F4) @LOG(A7)

C7: (F4) [W11] 1.6779

D7: (F4) [W9] (6310*(B7-B38)+C$8)

E7: (F4) [W9] (D312+C318*B7)

A8: 800

B8: (F4) @LOG(A8)

C8: (F4) [W11] 1.7587

D8: (F4) [W9] (GSlO‘(BS-B$8)+C$8)

E8: (F4) [W9] (D812+C$l8*B8)

A9: 1600

B9: (F4) @LOG(A9)

C9: (F4) [W11] 1.8578

12812838252!

111

D9: (F4) [W9] (6310*(B9-BS8)+C$8)
F9: [W16] "W METHOD 3

F10: [W16] 'Cvcc =

610: (F4) (C9-C8)/(B9-B8)

B11: 'Regressiorr Output:

E11: [W9] ’B.D8cc

F11: [W16] 'X =

611: (F4)

(DS 12+G$10*B$9-C$9)/(G$ 10-C318)

A12: ’Constant

D12: (F4) [W9] 1.4445859880058
F12: [W16] 'Log Pre pressu -

612: (F4) @LOG(G$l4)

A13: ’Std Err of Y Est

D13: (F4) [W9] 0.010651455299629
E13: (F4) [W9] (G$2‘(B3-B$4)+C$4)
A14: ’R Squared

D14: (F4) [W9] 0.91348565970866
E14: (F4) [W9] (652*(B4-BS4)+C$4)
F14: [W16] ’Prec. Pressure =

614: (F0) 10‘6311

A15: ’No. of Observations

D15: [W9] 4

E15: (F4) [W9] (682*(BS-BS4)+C$4)
F15: [W16] ’Bulk Density =

615: (F2) (133121-C8181'6312)

A16: 'Degrees of Freedom

D16: [W9] 2

E16: (F4) [W9] (632*(B6-B$4)+C$4)
E17: (F4) [W9] (632*(B7-B$4)+C$4)
A18: ’X Coefficient(s)

C18: (F4) [W11] 0.072717005997085
E18: (F4) [W9] (GSZ‘(B8-B$4)+C$4)

APPENDIX 2: Computer screen and cells of the free flow spreadsheet for

computation of the compression test.

112

 

8:88 88.: :8; 88.8 88.“ 888 28.8 88.8 38.8 82
38.8 88.8 55 28.8 88.8 88.8 88.8 48:. 4:3 8
$8.8 $8.2 83; 88.8 88.." 4.88.: 88.8 883 8:3 8..
88.: 8:8 88.. 8:34. 88." 88.8 88.8 88.8 3:3 8N
83.8 :85. :83 88.3. 88.8 88.8 88.8 28.8 88.8 82
28.“ 82.9. $92 88.8 22.8 8:8 885 88.8 82.8 8
«:2 38.8 8:... 88.8 84.3 8:8 28.8 88.8 88.8 8
83.2. 88.. 88.8 88.“ 8:8 88... o

s: as as use nee 88 oEe. 88 83 sec

zozbeomu 2.595.. cm 8233 .598: 88> o o m 4 <5. .25 885.8

8:8 "8 288.: u 82

8 Ba: "8.? 79.9. 8.8 0 Ease.

:8 8.8 15 2588.8 nee

so 83 "a: 2533 "an

a: e - _ 8.588.892 em: 2 N oz 382%

3.825.488

.25.. 5:8 ham—Em. 8:828 85 88:8 3 .55 588: 888: 85 8.555388 85 8.8.88 on. .5388.» b38538 8288»

85 E 88: 85 an 5838? 8e :8 .85 829cm 85 3852 5:8 .88. 8388.an8 85 Co 555358 85 «accrue 582.5888 5E.

Spreadsheet cells.

A1:
B1:
C1:
D1:
El:

ASAMPLE

2

“R2

’MSU
“NT-T-DEPTH

l-6kPa

A2:
B2:
C2:
D2:
E2:
F2:
A3:
B3:
C3:
D3:
E3:
F3:
A4:
B4:
C4:
D4:
E4:
F4:
A5:
B5:
C5:
D5:
E5:
A7:
B7:
C7:
D7:
E7:
F7:
G7:
H7:

'BDi=
1.38

’g cm-3
"Hs=

1.36

’cm

"D P=
2.65

’g cm-3
"Hi:

2.6

’cm

’Moist i=
(F2) 0.2496
’ks ks-l
"Ws=
113.81

’8

’Area =
31.67
’cm-2

IEi=
0.9118
“LOAD
“DIAL REA
“DELTA H
“DELTA E
“VOID
“HEIGHT
“VOLUME
“BD

17: ‘POROSITY

J7: [W11]
“REDUCTION

A8:
B8:
C8:

’ (KPa)
“ (cm)
“ (cm)

113

E8: “RATIO

F8: “ (cm)

G8: “ (cm)

H8: “ (g/cnr3)

18: “ (%)

J8: [W11] “ (96)

B9: ’

A10: (FO) 0

810: (F4) 0

E10: (F4) +E$5
F10: (F4) +1533
610: (F4) +F10*B$5
H10: (F4) +E$4lGlO
110: (F4) (1-H10/BS3)*100
All: (F0) 25

B11: (F4) 0.0564
C11: (F4) (Bll-BlO)
D11: (F4) +Cll/E$2
Ell: (F4) +E10-D11
F11: (F4) +F10—C11
611: (F4) +F11*B$5
H11: (F4) +E$4/Gll
111: (F4) (l-Hll/BS3)*100
J11: (F4) [W11]
(100-(111/1310)*100)
A12: (F0) 50

B12: (F4) 0.129

C12: (F4) (BIZ-Ell)
D12: (F4) +C12/E$2
E12: (F4) +E11-D12
F12: (F4) +F11-C12
612: (F4) +F12*B$5
H12: (F4) +E$4lGlZ
112: (F4) (1-Hl2/B$3)*100
J12: (F4) [W11]
(100-(112/ISlO)*100)
A13: (F0) 100

813: (F4) 0.2108
C13: (F4) (B13-B12)
D13: (F4) +C13lE$2
E13: (F4) +El2-Dl3
F13: (F4) +F12—C13
613: (F4) +F13*B$5

H13: (F4) +E34/Gl3
113: (F4) (1-H13/B33)"100
J13: (F4) [W11]
(100-(113/1310)*100)
A14: (F0) 200

B14: (F4) 0.3104
C14: (F4) (Bl4-B13)
D14: (F4) +C14/ES2
E14: (F4) +E13-Dl4
F14: (F4) +F13-C14
614: (F4) +F14‘B35
H14: (F4) +ES4/Gl4
114: (F4) (1—H14/BS3)*100
J14: (F4) [W11]
(100-(114/1310)*100)
A15: (PO) 400

B15: (F4) 0.414

C15: (F4) (B15-Bl4)
D15: (F4) +C15/ES2
E15: (F4) +El4-D15
F15: (F4) +Fl4-C15
G15: (F4) +F15*B$5
H15: (F4) +ES4/Gls
115: (F4) (1-H15/BS3)*100
J15: (F4) [W11]
(100-(115/1310)*100)
A16: (F0) 800

B16: (F4) 0.5144
C16: (F4) (Bl6-B15)
D16: (F4) +C16/ES2
E16: (F4) +E15-D16
F16: (F4) +F15-C16
G16: (F4) +Fl6*B$5
H16: (F4) +ES4/G16
116: (F4) (1-H16/BS3)*100
116: (F4) [W11]
(100-(Il6/L$10)*100)
A17: (F0) 1600

B17: 0.6114

C17: (F4) (B17-B16)
D17: (F4) +C17/ES2
E17: (F4) +El6—D17
F17: (F4) +F16—Cl7

114

617: (F4) +Fl7*B$5
H17: (1:4) +E$4lGl7

117: (F4) (l-H17/BS3)"'100
J17: (1:4) [W11]
(100-(117/1810)*100)

Ii till I i(|

nrcurGeN smrE UNIV. LIBReRIEs
lllllWINillilliiiilWill”"WWIWI
31293010466872