Terms and Conditions of Use

:
Visit our
companion site
http://www.vulcanhammer.org
this document downloaded from
vulcanhammer.net
Since 1997, your complete
online resource for
information geotecnical
engineering and deep
foundations:
The Wave Equation Page for
Piling
Online books on all aspects of
soil mechanics, foundations and
marine construction
Free general engineering and
geotechnical software
And much more...
All of the information, data and computer software
(“information”) presented on this web site is for general
information only. While every effort will be made to insure
its accuracy, this information should not be used or relied on
for any specifc application without independent, competent
professional examination and verifcation of its accuracy,
suitability and applicability by a licensed professional. Anyone
making use of this information does so at his or her own risk
and assumes any and all liability resulting from such use.
The entire risk as to quality or usability of the information
contained within is with the reader. In no event will this web
page or webmaster be held liable, nor does this web page
or its webmaster provide insurance against liability, for
any damages including lost profts, lost savings or any
other incidental or consequential damages arising from
the use or inability to use the information contained
within.
This site is not an offcial site of Prentice-Hall,
Pile Buck, the University of Tennessee at
Chattanooga, or Vulcan Foundation
Equipment. All references to sources of
software, equipment, parts, service
or repairs do not constitute an
endorsement.
G e ote xtile Re inforc e me nt of
Low- Be a ring - C a p a c ity Soils
C omp a rison of Two De sig n M e thod s
Ap p lic a b le to Tha wing Soils
S
P
E
C
I
A
L

R
E
P
O
R
T







9
9
-
7
Karen S. Henry June 1999
US Army Corps
of Engineers®
Cold Regions Research &
Engineering Laboratory
Abstract: Thawing fine-grained soils are often saturated
and have extremely low bearing capacity. Geosynthetics
are used to reinforce unsurfaced roads on weak, satu-
rated soils and therefore are good candidates for use in
stabilization of thawing soils. To stabilize the soil, a
geotextile is placed on it, then the geotextile is covered
with aggregate. Design involves selection of aggregate
thickness and geotextile. There are two commonly used
design techniques for geotextile reinforcement of low-
volume roads, and the Army uses one of them. The theory
and use of the two design methods for static loading
( i.e., up to 100 vehicle passes) are presented and com-
pared i n thi s report. The desi gn method not used
by the Army offers the potenti al to reduce aggregate
thi ckness over the geotexti l e because i t accounts for
the fact that the geotextile helps support the traffic load
( when in tension) and confines the soil between the
How to get copies of CRREL technical publications:
Department of Defense personnel and contractors may order reports through the Defense Technical Information Center:
DTIC-BR SUITE 0944
8725 JOHN J KINGMAN RD
FT BELVOIR VA 22060-6218
Telephone 1 800 225 3842
E-mail help@dtic.mil
msorders@dtic.mil
WWW http://www.dtic.dla.mil/
All others may order reports through the National Technical Information Service:
NTIS
5285 PORT ROYAL RD
SPRINGFIELD VA 22161
Telephone 1 800 553 6847 or 1 703 605 6000
1 703 487 4639 ( TDD for the hearing-impaired)
E-mail orders@ntis.fedworld.gov
WWW http://www.ntis.gov
A complete list of all CRREL technical publications is available from:
USACRREL ( CEERD-IM-HL)
72 LYME RD
HANOVER NH 03755-1290
Telephone 1 603 646 4338
E-mail techpubs@crrel.usace.army.mil
For information on all aspects of the Cold Regions Research and Engineering Laboratory, visit our World Wide Web site:
http://www.crrel.usace.army.mil
wheel s and the subgrade. However, thi s al ternati ve
method appears to be unconservative with respect to
stresses estimated at the subgrade surface. Thus, the
current Army desi gn techni que shoul d be used unti l
more research is conducted. In the meantime, straight-
forward design curves for Army 10- and 20-ton trucks
as well as vehicle loading and tire pressure informa-
ti on for a number of other vehi cl es are i ncl uded i n
this report to help make the current design method easy
to use.
Future work should consider adopting a hybrid de-
sign method that provides realistic estimates of stresses
at the subgrade and accounts for the tensile properties
of geotextiles. In addition, aggregates other than the high-
quality crushed rock that is inherently assumed by each
design method should be accounted for in new design
development.
G e ote xtile Re inforc e me nt of
Low- Be a ring - C a p a c ity Soils
C omp a rison of Two De sig n M e thod s
Ap p lic a b le to Tha wing Soils
Special Report 99-7
Karen S. Henry June 1999
Prepared for
OFFICE OF THE CHIEF OF ENGINEERS
Approved for public release; distribution is unlimited.
US Army Corps
of Engineers®
Cold Regions Research &
Engineering Laboratory
PREFACE
This report was written by Dr. Karen Henry, Research Civil Engineer of the Civil
Engineering Research Division, Research and Engineering Directorate, U.S. Army
Corps of Engineers Cold Regions Research and Engineering Laboratory, Hanover,
New Hampshire.
Funding for this work was provided by the U.S. Army Corps of Engineers under
Project/ Contract No. AT42, TO-007, Use of Geosynthetics to Rapidly Reinforce Thaw-
ing Soils.
The author thanks Professor Mark Evans of West Point Military Academy for
technical review of the manuscript. His review questions required responses that
added significantly to the breadth and depth of this work. She also thanks George
Blaisdell of CRREL for technical review as well as for numerous discussions of this
topic and encouragement. Finally, the contributions of Professor Robert D. Holtz of
the University of Washington are acknowledged. This work was initiated as an
independent study project while the author was a graduate student under his guid-
ance.
The contents of this report are not to be used for advertising or promotional
purposes. Citation of brand names does not constitute an official endorsement or
approval of the use of such commercial products.
ii
iii
CONTENTS
Page
Preface .......................................................................................................................... ii
Introduction ................................................................................................................. 1
Method currently used by the Army ....................................................................... 1
Design example ................................................................................................... 3
Theory .................................................................................................................. 4
Applicability for U.S. Army use on thawing soils ......................................... 6
Alternative method presented by Giroud and Noiray ......................................... 6
Design example ................................................................................................... 6
Theory .................................................................................................................. 9
Applicability for use by the U.S. Army ........................................................... 12
Comparison of Giroud and Noiray method with Army method ........................ 13
Validation of calculation techniques ................................................................ 13
Stress distribution through the aggregate layer ............................................. 13
Design curves for Army vehicles ..................................................................... 15
Recommendations for future work .......................................................................... 15
Summary and conclusions ........................................................................................ 21
Literature cited ............................................................................................................ 21
Appendix A: Geotextile survivability requirements ............................................. 23
Appendix B: Methods used to recalculate design curves from Barenberg
et al. and Giroud and Noiray ............................................................................ 25
Abstract ........................................................................................................................ 29
ILLUSTRATIONS
Figure Page
1. Aggregate thickness design curve for single-wheel load on gravel-
surface roads .................................................................................................... 2
2. Aggregate thickness design curve for dual-wheel load on gravel-
surface roads .................................................................................................... 2
3. Aggregate thickness design curve for tandem-wheel load on gravel-
surface roads .................................................................................................... 3
4. Wheel and axle configurations ........................................................................... 4
5. Design aggregate thickness chart for 88.95-kN wheel load ........................... 5
6. Aggregate thickness, h
o
’, and reduction of aggregate thickness, ∆h,
resulting from the use of geotextile as a function of soil cohesion
for 12-in. rut depth .......................................................................................... 7
7. Aggregate thickness, h
o
’, and reduction of aggregate thickness, ∆h,
resulting from the use of geotextile as a function of soil cohesion
for 8-in. rut depth ............................................................................................ 8
8. Diagram of “membrane effect” of geotextile reinforcement of thawing
soil, and shape of deformed geotextile ........................................................ 9
9. Definition of tire contact area for dual tires, and equivalent contact
area used in analysis ...................................................................................... 10
10. Wheel load distribution by the aggregate layer to the subgrade .................. 11
11. Static loading design curves from Barenberg et al. and design
points generated for this report according to method
documented in Appendix B .......................................................................... 14
12. Static loading design curves without geotextile from Giroud and Noiray
and points generated for this report according to method
documented in Appendix B .......................................................................... 14
13. Static loading design curves with geotextiles from Giroud and Noiray
and points generated for this report according to method
documented in Appendix B .......................................................................... 15
14. Static loading design curves adapted from Giroud and Noiray and
Barenberg et al. for 10- and 20-ton trucks with tire pressure of
414 kPa ............................................................................................................. 16
15. Field performers vs. theoretical prediction by Giroud and Noiray for
unreinforced test sections and reinforced test sections ............................. 17
16. Design curves for static loading for 10-ton dump truck, according
to Barenberg method and Giroud and Noiray method ............................ 18
17. Design curves for static loading for 20-ton dump truck, according
to Barenberg method and Giroud and Noiray method ............................ 19
TABLES
Table Page
1. Traffic loading data for Army vehicles .............................................................. 6
2. Thickness of plastic zone in the subgrade for dual-tired truck loading
and aggregate layer of thickness 0.15 m ...................................................... 11
3. Tensile modulus values of geotextiles at 5% strain and at failure
based on information in Geotechnical Fabrics Report .............................. 13
4. Maximum vertical stress at various depths below applied wheel load
of 115 kN and contact pressure of 414 kPa according to Newmark
and trapezoidal stress distribution used by Giroud and Noiray ............ 16
iv
Geotextile Reinforcement of Low-Bearing-Capacity Soils
Comparison of Two Design Methods Applicable to Thawing Soils
KAREN S. HENRY
INTRODUCTION
Thawing fine-grained soils are often saturated
or even supersaturated and thus have extremely
low bearing capacity. Geotextiles have been used
in the construction of low-volume, unsurfaced
roads on weak and saturated soils to reinforce the
base course–subgrade interface and therefore are
good candidates for use in stabilization of thaw-
ing soils. To stabilize weak soil with a geotextile
for trafficking, the geotextile is placed directly on
the soil and then covered with aggregate. The de-
sign involves selecting aggregate thickness and the
geotextile. There are two commonly used tech-
niques for designing for soil reinforcement using
geotextiles, one of which is prescribed in U.S.
Army guidance.
The current Army design technique for static
loading of low-volume roads on low-bearing-ca-
pacity soils was examined for ease of use and ap-
plicability to the reinforcement of thawing soil,
specifically for Army vehicles. Static loading is
defined as up to 100 passes of a vehicle at the maxi-
mum wheel load and a minimum rut depth of 0.10
m (4 in.). Information about Army vehicle load-
ing and design curves for specific Army vehicles
is provided in this report to help make the design
technique easier to use.
Another design method that offers the poten-
tial to reduce required aggregate thickness over
the geotextile (and thus cost) was compared with
the Army method. Theory and results from both
design methods are presented in this report. Al-
though both design methods include traffic load-
ing for up to 1000 vehicle passes, here we deal only
with design for static loading, which involves a
maximum of 100 vehicle passes.
METHOD CURRENTLY USED BY THE ARMY
(TM 5-818-8)
The design method currently used by the U.S.
Army for the stabilization of low-bearing-capac-
ity soils for low-volume roads and trails with
geotextile and aggregate was developed by the
U.S. Forest Service (Steward et al. 1977) based on
theory presented by Barenberg et al. (1975).* The
design method is presented in TM5-818-8 (1995)
as a series of soil strength vs. aggregate thickness
design curves for various wheel loads (defined
below), with a tire pressure of 552 kPa (80 psi) (Fig.
1 through 3). The design procedure includes
1) Converting soil strength to an equivalent cohe-
sion, c.
2) Selecting a maximum wheel load.
3) Selecting a value for a bearing capacity factor,
N
c
. N
c
values used with design curves for static
loading are 6.0 with geotextile and 3.3 without
geotextile.
4) Using the product cN
c
in the appropriate de-
sign chart (e.g., Fig. 1), and determining the
depth of aggregate required with and without
a geotextile.
5) Determining which section is less costly to build.
6) If use of a geotextile is advantageous, specify-
ing one according to geotextile construction sur-
vivability requirements.
Although TM5-818-8 does not specify the ag-
gregate properties required for low-volume roads,
*The low-bearing-capacity soils were assumed to be soft,
cohesive soils by Barenberg et al. (1975).
2
aggregate that meets base course requirements is
presumably required. Field Manual FM5-430-001
(1994) recommends that base course material have
minimum CBR values of 80 to 100. Soils that yield
these values include crushed rock, mechanically
stabilized aggregates, and well-graded gravel (e.g.,
FM5-430-001; Holtz and Kovacs 1981). Further
evidence for the requirement of high-quality ag-
gregate is the fact that the design technique is
based on experiments that used crushed-rock ag-
gregate (Barenberg et al. 1975).
Guidance for selecting wheel loads and contact
pressures to use with this design is not given in
TM5-818-8; thus, it is now provided. For single and
cN
c
psi 10 100 1 50 5
kPa
10 100 500 50
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
70
60
50
40
30
0
in.
20
10
0
1.5
1.0
0.5
m
89.0 kN (20,000 lb)
66.7 kN (15,000 lb)
44.5 kN (10,000 lb)
22.2 kN ( 5,000 lb)
Wheel Load, W
P = 2W
kPa
0
1.5
1.0
0.5
m
cN
c
psi 10 100 1 50 5
10 100 500 50
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
70
60
50
40
30
0
in.
20
10
71.2 kN (16,000 lb)
89.0 kN (20,000 lb)
53.4 kN (12,000 lb)
40.0 kN ( 9,000 lb)
17.8 kN ( 4,000 lb)
P = 2W
Wheel Load, W
Figure 1. Aggregate thickness design curve for single-wheel load on gravel-surface roads.
(From TM 5-818-8.)
Figure 2. Aggregate thickness design curve for dual-wheel load on gravel-surface roads. (From
TM 5-818-8.)
3
dual wheels on a single axle (Fig. 4a and 4b), the
wheel load, W, is defined as the total load on ei-
ther the left or right side of the axle. The axle load,
P, a quantity used in other design methods, is de-
fined as the total load on the axle, or 2W. For tan-
dem axles (Fig. 4c and 4d), estimations of wheel
and axle loads vary. Barenberg et al. (1975) use a
wheel load of 0.66 times the total load on one side
of both of the tandem axles. Giroud and Noiray
(1981) obtained the design axle load by multiply-
ing the sum of the two axle loads by 0.60. Most
U.S. states allow maximum loads on each pair of
tandem axles equal to 0.563 times the maximum
allowable single axle load (Yoder and Witczak
1975). Contact pressures for use in design are ap-
proximately 0.9 to 1.0 times the tire inflation pres-
sure for single-tired vehicles and 0.70 to 0.75 times
the tire inflation pressure for dual tires (Barenberg
et al. 1975). However, for the design method pre-
sented in TM5-818-8, there is negligible difference
in the aggregate thickness design curves for ac-
tual tire inflation pressure vs. contact pressure (e.g.,
Fig. 5).
Design example
Given an 80-kN (18,000-lb) maximum expected
single axle load (40-kN or 9,000-lb wheel load) on
a dual-tired vehicle, determine the aggregate thick-
ness required with and without geotextile for a
soil cohesion of 52 kPa (7.5 psi) or CBR of 2.*,†
There will be approximately 100 passes of this
vehicle, and a rut depth of 0.3 m (12 in.) can be
tolerated. The tire inflation/ contact pressure is
equal to 552 kPa (80 psi).
1. Calculate cN
c
as (52)(3.3) = 172 kPa (25 psi) with-
out geotextile, and (52)(6) = 312 kPa (45 psi) with
geotextile.
2. Enter Figure 2 with a cN
c
value of 172 kPa (25
psi) for a wheel load of 40 kN (9,000 lb) to ob-
tain a value of 0.32 m (13 in.) of aggregate re-
quired without geotextile. Using a cN
c
value of
312 kPa (45 psi) for the same wheel load, 0.20 m
(8 in.) of aggregate is required with geotextile.
3. Determine whether the cost of the geotextile
exceeds the cost of 0.12 m (5 in.) aggregate,
which would be saved by using the geotextile.
4. If it is advantageous to use a geotextile, specify
one using Tables 2-2 through 2-4 in TM5-818-8
(based on the need for the geotextile to survive
cN
c
psi
0
1.5
1.0
0.5
m
10 100 1 50 5
kPa
10 100 500 50
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
70
60
50
40
30
0
in.
20
10
P =
2W
a
77.8 kN (17,500 lb)
166.8 kN (37,500 lb)
194.6 kN (43,750 lb)
106.8 kN (24,000 lb)
35.6 kN ( 8,000 lb)
Wheel Load, W*
a = 0.66 (Barenberg et al. 1975)
*W = Sum of load on all four
wheels on one side of
the axle
x
Figure 3. Aggregate thickness design curve for tandem (dual)-wheel load on gravel-surface
roads. (From TM 5-818-8.)
*The relationship between shear strength (cohesion) or
CBR and Cone Index is given in Figure 2-3 of TM5-818-
8 (1995).
†Saturated silts and clays are likely to have CBR values
of this order of magnitude. Thawing frost-susceptible
soils may have even lower CBR values.
4
construction). Alternatively, the guidance pre-
sented in Appendix A, which was developed
more recently than that listed in TM5-818-8
(1995), can be used to select a geotextile.
Theory
Bender and Barenberg (1978) summarize the
theory and tests that led to the design method
described above. Using the Mohr–Coulomb fail-
ure criteria for soils, the shear strength is


s c p · + tan φ (1)
where s is the shear strength of the soil, c is the
cohesion, is the effective stress, and φ is the angle
of internal friction. For soft clay subgrades at or
near saturation, moving wheel loads are transient,
meaning that undrained loading applies. Thus, the
angle of internal friction is zero and the undrained
shear strength of the soil is equal to its cohesion.
Based on the theory of plastic equilibrium, the ul-
timate bearing capacity, q
d
, for soil in this condi-
tion is
q
d
= (2 + π)c . (2)
e
(b) Dual Wheels, Single Axle
(c) Dual Wheels, Tandem Axles
(d) Single Wheels, Tandem Axles
(a) Single Wheel, Single Axle
e
e
l = 1.8 to 2.0 m
e
l = 1.8 to 2.0 m
Figure 4. Wheel and axle configurations.
p + t
5
However, localized plastic strains that can cause
localized shear failure begin at
q ≈ πc . (3)
Barenberg et al. (1975) conducted laboratory
tests (two dimensional, cyclic loading) with a
geotextile (Mirafi 140) placed between crushed-
rock aggregate and a saturated clay subgrade.
Stress levels on the subgrade were estimated by
using a Boussinesq stress distribution beneath a
circularly loaded area (e.g., Newmark 1942), and
ratios between the calculated subgrade stress and
measured soil strength were developed. The al-
lowable stress with geotextile on the subgrade
surface was
σ
zallowable
= 6c . (4)
However, without geotextile, this relationship was
σ
zallowable
= 3.3c . (5)
These numbers are very close to the theoretical
values of general and local bearing capacity fail-
ure (eq 2 and 3). In addition to the change in fail-
ure mode from local to general bearing capacity
failure, the soil systems that contained geotextile
reached a level of permanent deformation so that
further loading of the same magnitude caused
negligible additional deformation. The unre-
inforced systems deformed progressively with
repeated loading.
Barenberg et al. (1975) constructed design charts
for aggregate thickness vs. soil strength by assum-
ing that the allowable pressure at the subgrade is
3.3c without geotextile and 6c with geotextile (Fig.
5). Stress at the subgrade was calculated by using
Boussinesq stress distribution beneath a circularly
loaded area (Newmark 1942). The contact area, A,
was determined by dividing the wheel load by the
contact pressure. The radius, r, needed for deter-
mination of the stress at the subgrade surface, was
obtained from A = πr
2
.
Barenberg et al. (1975) did not consider tensile
modulus or strength (or any mechanical property)
of t he geot ext ile in d eveloping t heir d esign
method. Furthermore, even though Bender and
Barenberg (1978) note that “a layer of aggregate
material is always needed on top of the fabric to
anchor it so that the necessary tensile forces can
be developed in the fabric” (p. 66), neither the
minimum depth for anchorage nor the mechani-
cal properties of the aggregate layer are specified in
eit her Barenberg et al. (1975) or Bender and
Barenberg (1978). The effects of traffic loading when
vehicle passes exceed 100 were accounted for by
Steward et al. (1977) by reducing the N
c
values.
50 40 30 20 10 0
psi
kPa
0 0.25 0.50 0.75 1.00 1.25 1.50
CBR
Without Geotextile
With Geotextile
Contact Pressure = 55.2 kPa ( 80 psi)
Contact Pressure = 103.4 kPa (150 psi)
0
1.2
0.6
0.4
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
50
20
0
m in.
40
30
10
1.0
0.8
0.2
0 7 1 2 3 4 5 6
Shear
Strength
Undrained
Figure 5. Design aggregate thickness chart for a 88.95-kN (20,000-lb) wheel load. (From
Barenberg et al. 1975.)
6
Applicability for Army use on
thawing soils
The design curves supplied in TM5-818-8 (Fig.
1 through 3) apply to vehicles with a 552 kPa
(80 psi) tire pressure, for single and dual wheel
loads varying from 22.24 to 88.96 kN (5,000 to
20,000 lb). Since TM5-818-8 does not list typical
wheel loads for Army vehicles (nor does U.S.
Army Field Manual FM5-430-00-1), some are pro-
vided for vehicles most likely to be used on
low-volume roads or trails (Table 1). The maxi-
mum wheel load for tandem axles listed in Table
1 was determined by multiplying the total load
on t he r ear axles by 0.60 (e.g., Girou d and
Noiray 1981) then dividing by two. For the 20-ton
dump truck (three rear axles), the maximum wheel
load was est imat ed by mult iplying t he t ot al
load on the rear axles by 0.70 then dividing it
by two.
The maximum wheel loads listed in Table 1 for
Army vehicles are reasonably well-represented in
Figures 1 through 3. Even though the tire pres-
sures in the figures are higher than the tire pres-
sures listed in Table 1, Barenberg et al. (1975) dem-
onstrated negligible difference in the design curves
due to variation in the contact pressures ranging
from 552 to 1034 kPa (80 to 150 psi). This theory
and design method assume that the subgrade is
uniform and that full plastic failure zones can de-
velop, the depth of which depends on the geom-
etry and magnitude of the loading. Thus, this
method was not intended for shallow thaw lay-
ers, and it would be conservative to use this
method for shallow thawed layers. Bounds on the
depth of the thawed layer for full development
of the plastic zone are discussed in the following
section.
ALTERNATIVE METHOD PRESENTED BY
GIROUD AND NOIRAY
Giroud and Noiray (1981) developed a design
method for geotextile placement between the ag-
gregate and subgrade of unpaved roads based on
bearing capacity theory for static loading. The
method accounts for the load support and soil
confinement provided by the geotextile itself and
is presented as a set of curves for dual wheels on a
single axle, with an axle load of 80 kN (18,000) at
various rut depths (Fig. 6 and 7). The curves are
used to determine aggregate thickness without
geotextile (h
o
′) and reduction of aggregate thick-
ness with geotextile (∆h) for geotextiles with dif-
ferent values of tensile modulus, K. The Giroud
and Noiray (1981) design method was chosen for
comparison with the method now used by the U.S.
Army because it is widely used (e.g., Holtz et al.
1995) and because of its potential for cost savings
by allowing thinner aggregate layers over the
geotextile because it takes into account the tensile
properties of the geotextile.
Design example
Given the same vehicle and soil conditions de-
scribed in the previous design example for the
method currently used by the Army, determine the
aggregate thickness required with a geotextile of
modulus K of 100 kN/ m (570 lb/ in.) and without
a geotextile. Conditions: 80 kN load on a dual-
wheel single axle, soil cohesion of 52 kPa (7.5 psi),
approximately 100 passes of the vehicle, tire infla-
tion pressure of 552 kPa (80 psi), and a tolerable
rut depth of 0.3 m (12 in.). Although Figure 6 is
constructed for a tire inflation pressure of 480 kPa
(70 psi), little difference in aggregate thickness is
Table 1. Traffic loading data for Army vehicles (Foss 1983).
Gross vehicle Load on Maximum Tire Ground
Vehicle/axle weight rear axles wheel load pressure clearance
type (kN/lb) (kN/lb) (kN/lb) (kPa/psi) (m/in.)
HMMWV 33.31/ 7,489 21.65*/ 4,869 10.83/ 2,434 241/ 35 0.41/ 16
M939 (6 × 6) 5-ton cargo truck 146.80/ 33,000 138.59/ 31,156 41.58/ 9,346 345/ 50† 0.30/ 12
HEMTT, M985 302.48/ 68,000 169.0/ 38,000 50.7/ 11,400 483/ 70† 0.30/ 12
M125 10-ton truck 289.14/ 65,000 187.94*/ 4,225 56.4/ 12,680 Not available 0.52/ 20
Articulated 8 × 8 Not available 258.0/ 58,000 77.4/ 17,400 414/ 60 0.30/ 12
M917 20-ton dump truck 324.30/ 72,906 324.30/ 72,906 113.51/ 25,520 414/ 60 Not available
Notes:
*No rear axle load was given; it was assumed that 65% of the gross vehicle load is applied on the rear axles.
†From Jeffrey Stark (personal communication, CRREL, 1997).
7
expected (e.g., Barenberg et al. 1975, Giroud and
Noiray 1981).
1. Enter the top chart in Figure 6 with a soil cohe-
sion value, c, of 52 kPa, for 100 vehicle passes
and determine that h
o
′ is 0.20 m (8 in.) of aggre-
gate required without a geotextile.
2. Enter the bottom chart in Figure 6 with a soil
cohesion value of 52 kPa, for 100 vehicle passes,
and for a geotextile with K = 100 kN/ m and de-
termine that the reduction in aggregate thick-
ness ∆h allowed is 0.12 m (5 in.). Thus, the ag-
gregate thickness theoretically allowed when a
geotextile is used is 0.06 m (3 in.). However, the
KH-78
120 100 80 60 40 20 0 kPa
1.0
0
0.8
0.6
0.4
0.2
m
0 1 2 3 4
CBR
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
,

h
o
35
30
25
20
15
0
in.
10
5
0 14 2 4 6 8 10 12 psi 16
Shear
Strength
Undrained
Axle Load = 80 kN (18,000 lb)
Rut Depth = 0.3 m (12 in.)
Tire Pressure = 480 kPa (70 psi)
N = 100
N = 10
1.0
0
0.8
0.6
0.4
0.2
120 100 80 60 40 20 0 kPa
m
0 1
Axle Load = 80 kN (18,000 lb)
Rut Depth = 0.3 m (12 in.)
Tire Pressure = 480 kPa (70 psi)
140
D
e
c
r
e
a
s
e

i
n

A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
,


h 35
30
25
20
15
0
in.
10
5
1
2
3
K = 450 kN/m
K = 400 kN/m
K = 300 kN/m
4
5
6
K = 200 kN/m
K = 100 kN/m
K = 10 kN/m
6
5
4
3
1
2
0 16 18
Shear
Strength
Undrained
Figure 6. Aggregate thickness, h
o
′ (top), and reduction of aggregate thickness, ∆h (bottom),
resulting from use of geotextile as a function of soil cohesion for 12-in. rut depth. N is num-
ber of vehicle passes and K is tensile modulus of geotextile. (From Giroud and Noiray 1981.)
8
aggregate layer should probably be a minimum
of 0.10 to 0.15 m (4-6 in.) to prevent damage to
the geotextile as discussed below.
Notice that for the same conditions without
geotextile, less aggregate is recommended by the
Giroud and Noiray design method than by the
Army method—0.2 m (8 in.) compared with 0.25
m (10 in.). The difference results from how each
method estimates stress at the subgrade; this is
discussed in Theory, below, and in Stress Distribu-
tion Through the Aggregate Layer in the comparison
of the Giroud and Noiray and Army methods.
Although the total amount of aggregate required
by the two design methods is different, approxi-
mately the same aggregate savings are realized by
120 100 80 60 40 20 0 kPa
1.0
0
0.8
0.6
0.4
0.2
m
0 1 2 3 4
CBR
Axle Load = 80 kN (18,000 lb)
Rut Depth = 0.2 m (8 in.)
Tire Pressure = 480 kPa (70 psi)
N = 100
N = 10
35
30
25
20
15
0
in.
10
5
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
,

h
o
0 14 2 4 6 8 10 12 psi 16
Shear
Strength
Undrained
1.0
0
0.8
0.6
0.4
0.2
m
0
Axle Load = 80 kN (18,000 lb)
Rut Depth = 0.2 m (8 in.)
Tire Pressure = 480 kPa (70 psi)
120 100 80 60 40 20 0 kPa 140
1
2
3
D
e
c
r
e
a
s
e

i
n

A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
,


h 35
30
25
20
15
0
in.
10
5
0 16 18
Shear
Strength
Undrained
1 K = 50 kN/m
2 K = 200 kN/m
3 K = 350 kN/m
∆h for:
Figure 7. Aggregate thickness, h
o
′ (top), and reduction of aggregate thickness, ∆h (bottom),
resulting from use of geotextile as a function of soil cohesion for 8-in. rut depth. N is num-
ber of vehicle passes and K is tensile modulus of geotextile. (From Holtz and Sivakugan
1987.)
9
using the Giroud and Noiray method with a 100
kN/ m modulus geotextile and the Army method
(0.12 m vs. 0.13 m, or essentially 5 in. for each case).
If a geotextile with a modulus of 300 kN/ m
(1,715 lb/ in.) is used, however, Figure 6 indicates
that no aggregate is required since ∆h = 0.2 m (8
in.). However, a minimum 0.10 to 0.15 m (4 to 6
in.) aggregate cover over a geotextile is recom-
mended at all times. The aggregate protects the
geotextile from damage imposed by construction
traffic as well as degradation due to exposure to
ultraviolet light (sunlight). It also helps anchor the
geotextile to allow it to develop the required ten-
sion.
Theory
Giroud and Noiray (1981) assumed a soft, satu-
rated cohesive subgrade in undrained loading, and
that the effect of the geotextile placed between the
aggregate and the subgrade will change the bear-
ing capacity failure from local (near the elastic
limit, see e.g., Whitman and Hoeg 1965) to gen-
eral (plastic). Thus, they applied the same soil
mechanics principles as did Barenberg et al. (1975).
However, Giroud and Noiray (1981) extended this
concept to account for the “membrane effect” of
the geotextile. Membrane effect refers to the fact
that the material contained by the concave side of
a stretched, flexible membrane is at a higher pres-
sure that the material on the outside of it. As bear-
ing capacity failure deforms the subgrade, the
geotextile undergoes deformation that puts it in
tension. The tensile strength of the geotextile then
helps to both support the load and confine the soil
above the geotextile, making it stronger (Fig. 8).
The modulus, K, of the geotextile is increasingly
influential as the rut depths increase (membrane
action occurs at large strains).
Like that of Barenberg et al. (1975), this theory
and the design technique is based on the assump-
Aggregate
Subgrade Soil
Geotextile
e
Geotextile
r s s
B B A A
t t t t
2a 2a
e
β β β
β
Initial Location
of Geotextile
(P)
(P )
2a
Figure 8. Diagram of “membrane effect” of geotextile reinforcement of thawing soil (top),
and shape of deformed geotextile (bottom). (After Giroud and Noiray 1981.)
10
tion that the subgrade soil is of a sufficient depth,
H
min
, to allow the plastic zones associated with
ultimate bearing capacity to develop. For the stress
dist ribut ion assumed by Giroud and Noiray
(1981), this amounts to
(6)
where B is the width of the loaded area at the soil
surface (Fig. 9b), h is the thickness of the aggre-
gate layer, and α = (π/ 4) – (φ/ 2), where φ is the
friction angle of the base course expressed as ra-
dians (Fig. 10). Giroud and Noiray (1981) assumed
the value of tan α to be 0.6. Assuming a dual-tired
truck with an axle load, P, having tire pressures,
P
c
, the width, B (m), of the wheel load is given by
(Giroud and Noiray 1981)
for off-highway trucks and
for on-highway trucks. (7)
Table 2 presents values for H
min
for a minimal ag-
gregate cover of 0.15 m (6 in.).
Thus, the Giroud and Noiray (1981) technique
is not generally applicable to thawed (or weak)
layers less than 0.4 m (16 in.) thick, and the same
is assumed for the current U.S. Army design tech-
nique. If a geotextile is used and the full plastic
zones do not develop, the tension in the geotextile
will not be fully mobilized. However, assuming
that the subgrade soil underlying the thawed soil
is stronger than the thawing soil, the support re-
quired of the geotextile will also be less than if the
subgrade were uniformly weak; and, therefore (as
mentioned above), designing for reinforcement
with geotextile would be conservative. When the
thawed layer is so thin that full plastic zones can-
not develop, the geotextile may provide impor-
tant separation between the thawing soil and the
aggregate that will likely lead to longer use of the
road without maintenance.
Ot her assumpt ions per t aining t o t he geo-
synthetic include that
1. The geotextile does not fail,
2. The shape of the deformed geotextile consists
of parabolas (Fig. 8, bottom),
3. The aggregate will not slide along the geotextile
surface,
4. The elongation, or strain, is uniform along the
entire length of the geotextile, and
5. The modulus of the geotextile, K, used in de-
L L
B B
Aggregate
e
(a)
Soil mechanically
associated with
adjacent tires
A
c
(b)
Figure 9. Definition of tire contact
area for dual tires (top), and equiva-
lent contact area used in analysis
(Giroud and Noiray 1981) (bottom).
(For single tires, L and B refer to
length and width of single tire print,
respectively.)

H
B h
min
·
+ 2
2
tan α

B
P
P
c
·

B
P
P
c
·
2
sign is the secant modulus obtained from ten-
sile tests.*
The fir st assu mp t ion is r easonable for
geotextiles that meet survivability requirements.
Measurements of test sections that were carefully
trafficked to minimize the wander of wheels in
lanes indicated that the shape of the deformed
geotextile is approximately parabolic (e.g., Kinney
and Barenberg 1979, Fannin and Sigurdsson 1996).
If the geotextile–aggregate friction is inadequate,
the aggregate can (and most likely does) slide
along the surface of the geotextile (e.g., Kinney and
Barenberg 1979, Kinney 1982); however, use of a
high-qu alit y aggregat e (as assu med by t his
method) would probably prevent this from occur-
ring. Tests have also shown that the strain is not
complet ely unifor m along t he lengt h of t he
geot ext ile (e.g., Kinney and Barenberg 1979,
Fannin and Sigurdsson 1996). For example, Fannin
and Sigurdsson (1996) measured the strain in three
geotextiles placed beneath a 0.25-m-thick layer of
sandy gravel and over a subgrade of average
st r engt h of 40kPa. The st r ain in t he t hr ee
geotextiles at 80 passes of a standard axle load of
80 kN and 620 kPa tire pressure averaged 0.9, 2.1,
and 5.2%, while the maximum strain in each of
the geotextiles was 2.6, 3.3, and 7.4%, respectively.
The influence of the inaccuracy of this assump-
tion on the validity of the design technique is not
yet clear.
It is also not clear what modulus value should
be used with this design method. Giroud and
Noiray (1981) recommend the use of a biaxial ten-
sile test, where the lateral deformation of the
geotextile is prevented during testing, and that the
secant modulus in the transverse direction of the
road be used. Biaxial testing would lead to esti-
mates of modulus values that are higher than those
determined in uniaxial tests by about 1.1 to 1.35
times (e.g., Giroud 1992, Soderman and Giroud
1995). However, Kinney (1982, 1998, personal com-
11
2a
B
L
Aggregate
Surface
h
Geotextile
γh
Subgrade/Aggregate
Interface
α
Figure 10. Wheel load distribution by aggregate layer to subgrade (Giroud and Noiray 1981).
Table 2. Thickness of plastic zone in the subgrade
for dual-tired truck loading and aggregate layer
of thickness 0.15 m (6 in.).
Axle load Plastic zone Plastic zone
kN (lb)/ thickness, on- thickness, off-
tire pressure highway truck highway truck
kPa (psi) H
min
(m/in.) H
min
(m/in.)
80 kN (18,000 lb)/ 0.42/ 16.5 0.47/ 18.5
480 kPa (70 psi)
60 kN (13,500 lb) 0.38/ 15 0.43/ 17
480 kPa (70 psi)
*The tensile modules of geotextiles can now be obtained
from ASTM D 4595 (1998) Standard test method for ten-
sile properties of geotextiles by the wide-width strip
method.
γh
munication) claimed that repeated loading of the
only geotextile that he tested resulted in an effec-
tive modulus that was “many times lower” than
t hat d et er mined from monot onically load ed
uniaxial and biaxial tests. Furthermore, in typical
stress–strain relations for tensile loading of needle-
punched geotextiles, the slope of the stress–strain
curve is initially quite low, resulting in low modu-
lus values at low strains. Therefore, research to
determine the effective modulus values when
geotextiles are being repeatedly loaded or traf-
ficked in-situ would be useful.
In addition to including the tensile support pro-
vided by the geotextile, there are two more ways
in which the Giroud and Noiray theory differs
from that presented by Barenberg et al. (1977).
Most significant is the shape of the stress distribu-
tion through the aggregate layer to the subgrade.
Giroud and Noiray (1981) used a trapezoidal dis-
tribution of the stress beneath a loaded rectangle
(Fig. 10) as opposed to the Boussinesq distribu-
tion beneath a circular plate used by Barenberg et
al. (1975). The assumed shape of the load and the
assumed stress distribution through the aggregate
layer to the subgrade results in significant differ-
ences in the estimated stresses at the subgrade for
certain loading and soil conditions. The difference
is especially significant for relatively thin aggre-
gate layers (less than approximately 0.3 m or 12
in.), as will be demonstrated in the next section.
Giroud and Noiray (1981) also assumed a mini-
mum CBR value of 80 for the overlying aggregate,
but Barenberg et al. (1975) did not discuss the
mechanical properties of aggregate, although the
tests they performed utilized crushed-rock aggre-
gate.
In addition to eq 8, the design equations from
Giroud and Noiray are
a) for off-highway trucks and
for on-highway trucks (9)
where L is the length of the rectangle formed by a
set of dual wheels (Fig. 9).
b) 2a = B + 2h tan α (10)
where a is defined in Figure 8b.
c) 2a’ = e – B – 2h tan α (11)
12
where e and a′ are defined in Figure 8 (bottom).
d)

s
ra
a a
·

+ ′
for a > a´ and
for a > a´ (12)
where s and r are defined in Figure 8 (bottom).
Applicability for use by the Army
The design curves based on the Giroud and
Noiray (1981) method currently published are for
standard 80-kN (18,000-lb) axle loads for on-high-
way trucks with tire inflation pressures of 480 and
620 kPa (70 and 90 psi) (e.g., Fig. 6 and 7). Esti-
mated axle loads for U.S. Army vehicles range up
to 324 kN (73,000 lb), and tire pressures can be as
low as 241 kPa (35 psi) (Table 1). In addition, con-
sideration should be given to the shape of the
wheel load applied on the surface. Giroud and
Noiray (1981) assumed a single-axle dual-wheel
configuration, whereas many Army vehicles have
single tires on tandem axles (e.g., the HEMTT).
Thus, if this design technique were to be adopted
by the Army, design curves for Army vehicles
should be developed for higher axle/ wheel loads
and for variations in the shape of the applied
loading.
The method published by Giroud and Noiray
(1981) uses geotextile tensile modulus values rang-
ing from 10 to 450 kN/ m. Geotextile modulus val-
ues at 5% strain, provided by the manufacturers
for a variety of geotextiles, are presented in Table
3. Based on limited field experiments, 5% strain
appears to be a reasonable estimate for static load-
ing of geotextiles performing reinforcement over
low-bear ing-capacit y soils (e.g., Fannin and
Sigurdsson 1996). Table 3 indicates that the ten-
sile modulus values in the cross-machine direc-
tion, the direction that would be transverse to traf-
fic, of some products commercially available today
are significantly greater than those for which
Giroud and Noiray (1981) provided design curves.
This suggests that the design method should in-
clude higher modulus values. However, recall
from the above discussion that modulus values
are higher in biaxial tension, but possibly far lower
for repeat ed loading t han for monot onically
loaded uniaxial tests. In reinforcing low-bearing-
capacity soil, the geotextile is expected to undergo
both biaxial tension and repeated loading. There-
fore, field or other experimental work is needed
to help establish the effective modulus values of

s
ra
a aa a
·
+ ′ ′
2
2 3
2
2 2


L
B
·
2

L
B
·
2
the geotextiles when they are being used and to
relate them to the values measured in tensile tests.
COMPARISON OF GIROUD AND NOIRAY
METHOD WITH ARMY METHOD
The tensile reinforcement advantages offered
by high-strength geotextiles may offset the in-
creased cost. Therefore, the currently used Army
design technique is compared with the design
technique of Giroud and Noiray (1981) in this sec-
tion. Design curves provided in Barenberg et al.
(1975) and Giroud and Noiray (1981) for static
loading were reconstructed to verify that the cal-
culation techniques used for this work are accu-
rate. Design curves for the loading imposed by
typical military vehicles using each design method
are also presented to demonstrate potential aggre-
gate savings by use of the Giroud and Noiray
(1981) method.
Validation of calculation techniques
Design equations were programmed using
Mathcad 6.0 (Mathsoft 1995) to generate design
curves. Details are given in Appendix B. Figure 11
shows t he st at ic load d esign cu r ves fr om
Barenberg et al. (1975) and points calculated for
this work to verify the calculations. Similarly, Fig-
ures 12 are 13 are static load design curves from
Giroud and Noiray (1981) and from calculations
performed for this work. There is a difference be-
tween the curves generated for Figure 13 and those
from Giroud and Noiray (1981) for the 450 kN/ m
geotextile being used for the 480 kPa tire pressure.
This difference is estimated to be about 10% at the
very lowest values of aggregate thickness. The
reason for this discrepancy is unknown.
Stress distribution through the aggregate layer
Figure 14 shows the soil strength vs. aggregate
thickness curves for both design techniques with-
out geotextiles for dual wheels on a single axle
with wheel loads of 60 and 115 kN (13,500 and
25,850 lb) and tire pressures of 414 kPa (60 psi).
These represent 10-ton and 20-ton trucks (e.g.,
Table 1). The Barenberg et al. (1975) method is more
conservative at these loading conditions, and this
stems from the load distribution assumptions per-
taining to the spreading of the load beneath the
wheels. Table 4 shows the maximum vertical stress
at various depths below the load for a wheel load
of 115 kN and contact pressure of 414 kPa using
the Boussinesq stress distribution beneath a cir-
cularly loaded area (i.e., Newmark 1942) and the
trapezoidal stress distribution beneath a rectan-
gular load used by Giroud and Noiray (1981).
Barenberg et al. (1975) used the Boussinesq
stress distribution because experimental and field
work of ot hers show t hat st ress dist ribut ion
13
Table 3. Tensile modulus values of geotextiles at 5% strain and at failure based on informa-
tion in Geotechnical Fabrics Report (1996).
Construction, K at 5% strain K at failure
mass/area (kN/m)/(lb/in.) (kN/m)/(lb/in.)
(g/m
2
)/
Product (oz/yd
2
) MD XD MD XD
Amoco 2044 W-PP, na 420/ 2400 760/ 4340 700/ 4000 875/ 5000
Carthage FX-400MF W-PP, 427/ 12 386/ 2206 456/ 2606 542/ 3098 783/ 4475
Contech C-300 W/ S-PP, 200/ 6 174/ 994 210/ 1200 306/ 1749 383/ 2186
Huesker Comtrac 800 W-PET, 1430/ 42 7200/ 41150 800/ 4572 7910/ 45206 667/ 3810
Linq GTF 550T W-PET, na 404/ 2309 404/ 2309 876/ 5006 876/ 5006
Linq GTF 1000T W-PET, na 1050/ 6000 1050/ 6000 1402/ 8012 1402/ 8012
Synthetic Industries W/ S-PP, 150/ 4 174/ 994 192/ 1097 233/ 1333 300/ 1715
Gtx. 200ST
Synthetic Industries W/ C-PP, 440/ 13 384/ 2195 454/ 2595 500/ 2858 583/ 3334
Gtx. 4 × 4
TNS W300 W-PP, 203/ 6 100/ 570 280/ 1600 290/ 1657 310/ 1772
USA Spantex 5710 K-PET, 2566/ 76 8000/ 45720 4000/ 22860 10000/ 57150 4167/ 23814
Webtec, TTHPG-50 W-PP, na 200/ 1143 220/ 1257 267/ 1524 260/ 1486
Webtec, TTHPG-57 W-PP 700/ 4000 700/ 4000 538.5/ 3078 487.5/ 2786
Notes: na = not available, W = woven, K = knitted, PP = polypropylene, PET = polyester, MD = machine direction, XD
= cross-machine direction.
14
50 40 30 20 10 0 kPa
0 0.25 0.50 0.75 1.00 1.25 1.50
CBR
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
50
20
0
in.
40
30
10
Wheel Load = 89.0 kN (20,000 lb)
Tire Pressure = 520 kPa (75 psi)
Without Geotextile
Wheel Load = 22.2 kN (5,000 lb)
Tire Pressure = 345 kPa (50 psi)
With Geotextile
0
1.2
0.6
0.4
m
1.0
0.8
0.2
0 7 1 2 3 4 5 6 psi
Shear
Strength
Undrained
Barenberg et al. (1975)
This Study
}
Figure 11. Static loading design curves from Barenberg et al. (1975) and design points generated
for this report according to method documented in Appendix B.
140 120 100 80 60 40 20 0 kPa
1.4
0
0.8
0.6
0.4
0.2
m
1.2
1.0
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s 40
30
20
10
0
in.
50
0 1 2 3 4
CBR
Axle Load = 870 kN (196,000 lb)
Tire Pressure = 620 kPa (90 psi)
Off-Highway Vehicle
Axle Load = 80 kN (18,000 lb)
Tire Pressure = 480 kPa (70 psi)
On-Highway Vehicle
0 14 2 4 6 8 10 12 psi 16 18 20
Shear
Strength
Undrained
Barenberg et al. (1975)
This Study }
Figure 12. Static loading design curves without geotextile from Giroud and Noiray (1981) and
points generated for this report according to method documented in Appendix B.
15
Figure 13. Static loading design curves with geotextiles from Giroud and Noiray (1981) and points
generated for this report according to method documented in Appendix B.
100 90 80 70 60 50 40 30 20 10 0 kPa
K = 10 kN/m, Tire Inflation Pressure = 621 kPa (90 psi)
K = 450 kN/m, Tire Inflation Pressure = 480 kPa (70 psi)
1.0
0
0.8
0.6
0.4
0.2
m
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
35
30
25
20
15
0
in.
10
5
0 1 2 3
CBR
0 14 2 4 6 8 10 12 psi
Shear
Strength
Undrained
Giroud and Noiray (1981)
This Study
}
140 120 100 80 60 40 20 0 kPa
1.4
0
0.8
0.6
0.4
0.2
m
1.2
1.0
Barenberg wheel load = 115 kN (25,900 lb)
Barenberg wheel load = 60 kN (13,500 lb)
Giroud & Noiray = 115 kN (25,900 lb)
Giroud & Noiray = 60 kN (13,500 lb)
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s 40
30
20
10
0
in.
50
0 1 2 3 4
CBR
0 14 2 4 6 8 10 12 psi 16 18 20
Shear
Strength
Undrained
Figure 14. Static loading design curves adapted from Giroud and Noiray (1981) and Barenberg et
al. (1975) for 10- and 20-ton trucks with tire pressures of 414 kPa (60 psi).
through a granular layer to the subgrade follows
the same pattern as that given by the Boussinesq
theory. Yoder and Witzak (1975) also refer to the
use of a Boussinesq distribution of stresses below
traffic loading for the purposes of pavement de-
sign. Indeed, mobility models also incorporate
Boussinesq stress distributions.* Although trap-
ezoidal stress distribution below rectangular-
shaped loads is commonly used in shallow foun-
dation design (e.g., Perloff 1975), Giroud and
Noiray (1981) did not cite other work that uses
trapezoidal stress distribution to estimate traffic
loading stresses through aggregate.
The significant difference in estimation of
stresses at the surface used by the two methods
warrants further investigation. There is limited
evidence suggesting that the Giroud and Noiray
(1981) method is unconservative for static load-
ing conditions in both reinforced and unreinforced
test sections when the aggregate layers are 0.25 to
0.50 m thick and the subgrade strength ranges
from 30 to 40 kPa (CBR of about 1.5) (e.g., Fannin
and Sigurdsson 1996; Fig. 15). The ratio of the trap-
ezoidal stress below a rectangle to the Boussinesq
stress below a circular plate for these loading con-
ditions ranges from 0.61 for a 0.25-m- (10-in.-) thick
aggregate to 0.78 for the 0.5-m- (20-in.-) thick ag-
gregate (Giroud and Noiray 1981, Newmark 1942).
Thus, until further investigation, use of the guid-
ance in TM5-818-8, which incor p or at es t he
Boussinesq stress distribution through the aggre-
gate, is recommended.
The aggregate quality significantly influences
the stress distribution through it (Herner 1955),
and this should not be discounted as a potential
factor in the observed unconservative design for
static loading by the Giroud and Noiray method
described above. For example, when a 45-kN (10-
kip) load was applied by an airplane tire at 690
kPa (100 psi), the vertical stress reaching the
subgrade through a 0.6-m- (24-in.-) thick layer of
sand was about twice that of the stress reaching
the subgrade through a layer of crushed limestone
(Herner 1955). McMahon and Yoder (1960) dem-
onstrated that, for compacted, crushed limestone
base rock layers ranging in thickness from 0.1 to
0.3 m (4 to 12 in.) and loaded with circular plates,
16
Table 4. Maximum vertical stress at various depths below
applied wheel load of 115 kN and contact pressure of 414
kPa according to Newmark (1942) and trapezoidal stress dis-
tribution used by Giroud and Noiray (1981).*
Stress Stress
Depth according to according
below trapezoidal to Boussinesq Ratio of the
applied stress (Newmark) trapezoidal
stress, z distribution method stress to the
(m) (kPa/psi) (kPa/psi) Boussinesq stress
0.1 275.4/ 39.9 400.1/ 58.0 0.69
0.2 198.1/ 28.7 342.1/ 49.6 0.58
0.3 151.0/ 21.9 265.7/ 38.5 0.57
0.4 120.4/ 17.5 210.0/ 30.5 0.60
0.5 99.7/ 14.5 151.1/ 21.9 0.66
0.6 85.2/ 12.4 116.2/ 16.9 0.73
0.7 74.8/ 10.8 91.2/ 13.2 0.82
0.8 67.3/ 9.8 73.0/ 10.6 0.92
0.9 61.9/ 9.0 59.6/ 8.6 1.04
1.0 57.9/ 8.4 49.4/ 7.2 1.17
*The Boussinesq method used to generate results in this report did not add
the pressure due to the weight of the overburden (= γ z) whereas the trapezoi-
dal method used did. The calculations were carried out in this manner to be
consistent with how the original researchers presented them. If the weight of
the overburden were added to the stresses estimated by the Boussinesq
method, the differences in stresses at depths of up to 1 m would be even
greater than those listed in Table 5.
*Personal communication, G.L. Blaisdell, Research Civil
Engineer, US Army Cold Regions Research and Engi-
neering Laboratory, Hanover, N.H., 1997.
Figure 15. Field performance vs. theoretical prediction by Giroud and Noiray (1981)
for unreinforced test sections (top) and reinforced test sections (bottom). (From
Fannin and Sigurdsson 1996.)
17
the pressure measured in compacted clay-soil be-
low the rock layers was reasonably approximated
by a Boussinesq distribution beneath circular
plates. Barenberg et al. (1975) based their theory
on tests that utilized “crushed stone aggregate,”
and, based on the work of McMahon and Yoder
(1960), a Boussinesq stress distribution through it
is a reasonable assumption for such an aggregate.
Unfortunately, Steward et al. (1977) did not de-
scribe the aggregate that was used in tests to vali-
date the Barenberg et al. (1975) design method.
The possibility of using Boussinesq stress dis-
tribution through the aggregate layer could be
added to the Giroud and Noiray (1981) design
technique. In addition, shapes other than a circu-
larly loaded area should be considered, and work
that examines stress distributions through aggre-
gates other than crushed rock should also proceed.
This would allow the confident use of design tech-
niques for relatively low-quality aggregate that
might be the only option for theater of operations
military construction.
15
10
5
0
B
a
s
e

C
o
u
r
s
e

T
h
i
c
k
n
e
s
s
,

h
0.5
0.4
0.3
0.2
0.1
0
m
0.5
0.4
0.3
0.2
0.1
0
1 10 100 1000
r
=
5

c
m

(
2

i
n
.
)
r
=
1
5

c
m

(
6

i
n
.
)
r
=
5

c
m

(
2

i
n
.
)
r
=
1
5

c
m

(
6

i
n
.
)
Field Trial:
r = 5 cm (2 in.)
r = 10 cm (4 in.)
r = 15 cm (6 in.)
Austin et. el.)
r = 5 cm (2 in.)
r = 10 cm (4 in.)
30 kPa
(4.4 psi)
Su =
40 kPa
(5.8 psi)
Su =
Field Trial:
r = 5 cm (2 in.)
r = 10 cm (4 in.)
r = 15 cm (6 in.)
r
=

5

c
m

(
2

in
.
)
r

=

1
5

c
m

(
6

i
n
.
)
in.
15
10
5
0
Austin et al.
Design curves for Army vehicles
Because the potential for aggregate and cost
savings is of interest to the U.S. Army, and the
Giroud and Noiray (1981) method shows prom-
ise for large savings over the current Army design
method, design curves for Army vehicles were
developed according to both methods for compari-
son of aggregate thickness required. Design curves
for the U.S. Army’s 10- and 20-ton trucks are pre-
sented in Figures 16 and 17, respectively. A geo-
textile tensile modulus of 200 kN/ m (1143 lb/ in.)
was used for Figures 16 and 17 (bottom) because
this value is easily obtained for commercially avail-
able products (Table 3).
Considerable aggregate savings can be realized
if the Giroud and Noiray (1981) method is used.
For the 10-ton truck, with a soil strength of 30 kPa
(4.4 psi), the aggregate savings for the Giroud and
Noiray (1981) method over the current Army
method is about 0.2 m (8 in.) with geotextile. For
the 20-ton truck at a soil strength of 40 kPa (5.8
psi), the aggregate savings for the Giroud and
Noiray (1981) method over the current Army
method is about 0.2 m (8 in.) with geotextile. Thus,
accounting for the tensile support provided by the
geotextile provides considerable advantages of
aggregate savings. It is important to remember that
the aggregate used with this method should have
a minimum CBR of 80 (Giroud and Noiray 1981).
RECOMMENDATIONS FOR FUTURE WORK
Using the Giroud and Noiray (1981) method as
it is presented herein may lead to unconservative
design and construction results because the stress
distribution through the aggregate layer to the
subgrade is less than the Boussinesq method,
which is widely accepted and well-supported.
However, it should be further investigated because
it promises large aggregate savings compared with
the current design method, due solely to its abil-
ity to account for tensile properties of the geotextile
reinforcement at large rut depths, a situation that
can be tolerated by military vehicles on thawing
soils. Depending on the outcome of an investiga-
tion of the stress distribution through the aggre-
gate layer to the subgrade, it may be worthwhile
to develop a hybrid design method that uses a
Boussinesq stress distribution through the sub-
grade with a membrane support mechanism as
presented by Giroud and Noiray (1981). For use
of a Boussinesq stress distribution, the load at the
surface is not necessarily best modeled as a rigid
circular area (as it is now). For example, the length
to width ratio for a HEMTT is estimated as L =
1.6B (e.g., Richmond et al. 1990). Thus, other
wheel-load geometries should be considered.
The Giroud and Noiray (1981) design method
indicates that the geotextile may be able to pro-
vide reinforcement with no aggregate on the sur-
face. As discussed earlier, this is not a currently
recommended practice because of the increased
risk of damage to the geotextile due to trafficking
and because of deterioration when exposed to
sunlight (e.g., Holtz et al. 1993). However, it is
potentially of great interest to the U.S. Army for
reinforcement of thawing soils, especially for ex-
pedient, temporary operations where ultraviolet
degradation due to exposure to sunlight is not a
consideration (e.g., less than 10 days of exposure)
and aggregate is not available. For this concept to
be implemented, the geotextiles would likely have
to be anchored in some way in order for the ten-
sile properties to fully develop and provide the
necessar y r einforcement . (Even t hou gh t he
geotextile is in a state of tension between the
wheels, the portion on the outside of each set of
wheels could easily slip into ruts formed by the
vehicles.)
An important factor in the adoption of the
Giroud and Noiray (1981) design method for use
is knowledge of the appropriate geotextile modu-
lus values. Geotextile modulus values at 5% strain
are readily available. Based on limited field experi-
ments, this appears to be a reasonable strain esti-
mate for static loading of geotextiles performing
reinforcement over low-bearing-capacity soils.
However, modulus values are higher in biaxial
tension, but possibly far lower for repeated load-
ing than for monotonically loaded uniaxial tests—
the tests that are now performed to determine
geotextile modulus values. In reinforcing low-
bearing-capacity soil, the geotextile is expected to
undergo both biaxial tension and repeated load-
ing. Therefore, field or other experimental work
is needed to help establish the effective modulus
values of the geotextiles when they are being used
and to related them to the values measured in ten-
sile tests. Finally, the tensile modulus values of
some commercially available geotextiles far exceed
those used in Figures 16 and 17 (bottom). Future
work should consider the use of available prod-
ucts with appropriately high modulus values. This
could result in substantial aggregate savings.
Regardless of whether the Giroud and Noiray
18
0 14 2 4 6 8 10 12 psi
100 90 80 70 60 50 40 30 20 10 0 kPa
1.0
0
0.8
0.6
0.4
0.2
m
With Geotextile
Without Geotextile
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
35
30
25
20
15
0
in.
10
5
Shear
Strength
Undrained
0 1 2 3 4
CBR
100 90 80 70 60 50 40 30 20 10 0
1.0
0
0.8
0.6
0.4
0.2
m
Without Geotextile
With 200 kN/m
Geotextile
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
35
30
25
20
15
0
in.
10
5
0 14 2 4 6 8 10 12
Shear
Strength
Undrained
0 1 2 3 4
CBR
Figure 16. Design curves for static loading (up to 100 passes) for 10-ton dump truck, according to
Barenberg (1975) method (top) and Giroud and Noiray (1981) method (bottom). Use of upper figure
is recommended until further research is conducted.
19
Figure 17. Design curves for static loading (up to 100 passes) for 20-ton dump truck, according to
Barenberg (1975) method (top) and Giroud and Noiray (1981) method (bottom). Use of upper figure
is recommended until further research is conducted.
100 90 80 70 60 50 40 30 20 10 0 kPa
1.0
0
0.8
0.6
0.4
0.2
m
With Geotextile
Without Geotextile
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
0
in.
35
30
25
20
15
0
10
5
0 1 2 3 4
CBR
0 14 2 4 6 8 10 12 psi
Shear
Strength
Undrained
0 14 2 4 6 8 10 12 psi
100 90 80 70 60 50 40 30 20 10 0 kPa
1.0
0
0.8
0.6
0.4
0.2
m
Without Geotextile
With 200 kN/m
Geotextile
A
g
g
r
e
g
a
t
e

T
h
i
c
k
n
e
s
s
35
30
25
20
15
0
in.
10
5
Shear
Strength
Undrained
0 1 2 3 4
CBR
20
(1981) design method is selected for use by the U.S.
Army, estimates of stress through a variety of
aggregates, not just crushed rock, should also be
completed so the design method can be adjusted
accordingly. More rounded material such as sand
and gravel has been shown to concentrate stresses
over a significantly smaller area on the subgrade
than crushed rock. Due to the likelihood that the-
ater of operations construction will be completed
with limited sources of high-quality aggregate, this
would be an important addition to the current
design method.
Finally, even though soils are usually only tem-
porarily in a weakened state when they thaw, they
will sometimes have to carry more than 100 ve-
hicles during thawing. Thus, a method that ac-
counts for repeated traffic loading is desirable, and
this should also be included in future development
efforts.
SUMMARY AND CONCLUSIONS
Table 1 may be used with the design curves pre-
sented in TM5-818-2 for convenience in using the
current Army geotextile reinforcement design
method. However, if 10- or 20-ton trucks are ex-
pected to exert the maximum wheel loads on thaw-
ing or other low-bearing-capacity subgrade soils,
Figure 16 or 17 (top) may be used, respectively.
These methods both require the use of high qual-
ity aggregate. If the thawed layer is less than 0.4
m (16 in.) thick, these methods are likely to be con-
servative. However, a geotextile separator will
probably still provide benefit to lengthen times be-
tween maintenance of the gravel surface.
If further research proves that the Giroud and
Noiray (1981) design method is adequate, consid-
erable aggregate savings for the U.S. Army would
be realized by using it. However, since the Giroud
and Noir ay (1981) d esign met hod may be
unconservative, it should not be used by the U.S.
Army until further study is completed. A hybrid
method, combining a Boussinesq stress distribu-
tion through the aggregate layer with a membrane
support mechanism as presented by Giroud and
Noiray (1981) might be an optimum design tech-
nique. When this approach is further developed,
it should also include determination of represen-
tative modulus values, the use of a variety of ag-
gregates, the shape of the wheel load, and repeated
traffic loading.
LITERATURE CITED
AASHTO-AGC-ARTBA Joint Committee: Sub-
committee on New Highway Materials, Task
Force 25 (1990) Guide specifications and test pro-
cedures for geotextiles. American Association of
State Highway and Transportation Officials, Wash-
ington, D.C.
Austin, D.N., and D.M. Coleman (1993) A field
evaluation of geosynthetic-reinforced haul roads
over soft fou nd at ion soils. In Proceedings,
Geosynthetics 93, Industrial Fabrics Association In-
ternational, St. Paul, Minn., p. 65–80.
ASTM D 4595-86 (1998) Standard test method for
tensile properties of geotextiles by the wide-width
strip method. 1998 Annual book of ASTM Stan-
dards, Vol. 4.09, Soil and Rock (II): D4943–latest;
Geosynthetics.
Barenberg, E.J. , J. Hales, and J. Dowland (1975)
Evaluation of soil–aggregate systems with MIRAFI
fabric. University of Illinois Report No. UILU-
ENG-75-2020, prepared for Celanese Fibers Mar-
keting Company.
Bender, D.A., and E.J. Barenberg (1978) Design
and behavior of soil–fabric–aggregate systems.
Transportation Research Record 671, p. 64–75.
Fannin, R.J., and O. Sigurdsson (1996) Field ob-
servations on stabilization of unpaved roads with
geosynthetics. American Society of Civil Engineers
Journal of Geotechnical Engineering, July, 122(7): 544–
553.
FM5-430-001/AFJPAM 32-8013, Vol. 1 (1994) Plan-
ning and design of roads, airfields, and heliports
in the theater of operations—Road design. Head-
quarters, Department of the Army, Department of
the Air Force, Washington, D.C.
Foss, C.F. (1983) Jane’s Military Vehicles and Ground
Support Equipment. Fourth edition, Alexandria, Va.
Giroud, J.P. (1992) Biaxial tensile state of stress in
geosynthetics. Geotextiles and Geomembranes, Vol.
11, p. 319–325.
Giroud, J.P., and L. Noiray (1981) Geotexile-rein-
forced unpaved road design. Proceedings, of the
American Society of Civil Engineers Journal of the
Geot echnical Engineering Division, Sept ember,
107(GT9): 1233–1254.
Kinney, T.C. (1982) Discussion of geotextile-rein-
for ced u p aved r oad d esign. Journal of t he
Geotechnical Division, American Society of Civil En-
gineers, December, 108(GT12): 1657–1658.
Herner, R.C. (1955) Effect of base course quality
on load transmission through flexible pavement.
21
22
In Proceedings, Highway Research Board, 1955, p.
224–233.
Holtz, R.D., and W.D. Kovacs (1981) An Introduc-
tion to Geotechnical Engineering, Englewood Cliffs,
N.J.: Prentice-Hall, Inc.
Holtz, R.D., and N. Sivakugan (1987) Design
charts for roads with geotextiles. Geotextiles and
Geomembranes, Vol. 5, p. 191–199.
Holtz, R.D., B.R. Christopher and R.R. Berg
(1995) Geosynt het ic design and const ruct ion
guidelines. Participant notebook, FHWA HI-95-
038, U.S. Department of Transportation, Federal
Highway Administration, Washington, D.C.
Mathsoft (1995) Mathcad PLUS 6.0. MathSoft Inc.,
Cambridge, Mass.
McMahon, T.F., and E.J. Yoder (1960) Design
of a pressure-sensitive cell and model studies
of pressure in a flexible pavement subgrade. In
Proceedings, Highway Research Board, 1960, p. 650–
682.
Newmark, N.M. (1942) Influence charts for com-
putation of stresses in elastic foundations. Univer-
sity of Illinois Bulletin, Vol. 40, No. 12.
Perloff, W. H. (1975) Pressure distribution and
settlement. Foundation Engineering Handbook, H.F.
Winterkorn and H.Y. Fang, eds. Chapter 4. New
York: Van Nostrand Reinhold, p. 148–196.
Soderman, K.L., and J.P. Giroud (1992) Relation-
ship between uniaxial and biaxial stresses and
strains in geosynthetics. Geosynthetics International,
2(2): 495–504.
Steward, J.E., R. Williamson and J. Mohney
(1977) Guidelines for use of fabrics in construc-
tion and maintenance of low-volume roads. US
Forest Service, Portland, Oregon, Division of En-
gineering.
TM5-818-8/AFJMAN 32-1030 (1995) Engineering
use of geotextiles. Headquarters, Departments of
the Army and the Air Force, Washington, D.C.
Yoder, E.J., and M.W. Witczak (1975) Principles of
Pavement Design, Second edition. New York: John
Wiley and Sons.
APPENDIX A: GEOTEXTILE SURVIVABILITY REQUIREMENTS
This appendix is provided as a convenience to readers, so that geotextiles meet-
ing survivability requirements may be specified for acquisition without referring
to TM5-818-8 (1995). It can be used in place of Tables 2-2, 2-3, and 2-4 in U.S. Army
TM5-818-8. The guidance provided here is taken from AASHTO-AGC-ARTBA Joint
Committee: Subcommittee on New Highway Materials, Task Force 25 (1990) and is
more recent guidance than that provided in TM5-818-8.
Table A-1. Construction survivability ratings.
Site soil CBR <1 1 to 2 >3
Equipment ground contact >350(50) <350(50) >350(50) <350(50) >350(50) <350(50)
pressure, kPa (psi)
Cover thickness, compacted
(mm/ in.)
a
100/ 4
b,c
NR NR H H M M
150/ 6 NR NR H H M M
300/ 12 NR H M M M M
450/ 18 H M M M M M
a
Maximum aggregate size not to exceed one half of compacted cover thickness.
b
For low-volume, unpaved roads (average daily traffic less than 200 vehicles).
c
Minimum cover thickness is limited to existing road bases and is not intended for use in new construc-
tion.
H = high, M = medium, NR = not recommended.
23
Table A-2. Geotextile physical property requirements for
survivability
a
<50% elongation/>50% elongation
b,c
Grab Puncture Trapezoidal
Required strength resistance tear strength
survivability (kN/lb) (kN/lb) (kN/lb)
level ASTM D 4632 ASTM D 4833 ASTM D 4533
H 270/ 180 100/ 75 100/ 75
M 180/ 115 70/ 40 70/ 40
Additional requirements Test method
Apparent opening size ASTM D 4751
1. <50% soil finer than 0.075 mm, AOS <0.6 mm
2. >50% soil finer than 0.075 mm, AOS > 0.3 mm
Permeability ASTM D 4491
K
geotextile
> K
soil
K
geotextile
= permittivity × nominal geotextile thickness
Ultraviolet degradation ASTM D 4355
At 150 hr exposure, 70% of strength retained for
all cases
Geotextile acceptance ASTM D 4759
a
Values shown are minimum roll average values.
b
Elongation as determined by ASTM D 4632.
c
Values of geotextile elongation do not imply the allowable consolidation
properties of the subgrade soil. These must be determined by a separate
investigation.
24
Table A-3. Recommended overlaps.
Soil strength Unsewn overlap Sewn overlap
(CBR) (mm/in.) (mm/in.)
<1 NR 229/ 9
1–2 965/ 38 203/ 8
2–3 762/ 30 76/ 3
>3 610/ 24 —
NR = Not recommended.
25
APPENDIX B: METHODS USED TO RECALCULATE DESIGN CURVES FROM
BARENBERG ET AL. AND GIROUD AND NOIRAY.
The design method presented by Barenberg et al. (1975) necessitated the solving
of equations for the aggregate thickness required, given a known range of soil co-
hesion. For Giroud and Noiray (1981), equations for cohesion were solved for a
given range of aggregate thickness. Mathcad 6.0 (Mathsoft 1995) was used to solve
these equations and generate design curves as well as to produce a symbolic equa-
tion for the aggregate thickness required for the Barenberg et al. (1975) approach.
SOLUTION FOR DEPTH OF AGGREGATE FOR
BARENBERG DESIGN METHOD
Barenberg et al. (1975) assumed that the stress transmitted to the subgrade sur-
face through the aggregate layer can be approximated by a Boussinesq stress distri-
bution through an elastic, homogeneous, isotropic half-space. The ratio of vertical
stress at depth, z, to the stress on a uniformly loaded circular area in an elastic,
homogeneous, isotropic solid bounded by a plane horizontal surface is (Newmark
1942)
σ = 1 – cos
3
α (B.1)
where

α · ( ) a r z tan / , and r is the radius of the circle and z is the depth at which
the stress determination is desired (z is located directly below the center of the
circularly loaded area).
The expression for α is substituted into eq B.1 and is rearranged to yield

a
r
z
a tan cos – .
|
.

`
,
· ( ) 1
1
3
σ (B.2)
This equation is then solved for z (using Mathcad 6.0) to yield

z r ·
( )
( )
|
.

`
,

1
1 1
1
3
2
3

– –
.
σ
σ
For a range of soil cohesion, c, and a known applied stress, the stress that can be
tolerated at the subgrade is given by either 3.3 c (without geotextile) or 6.0 c with
geotextile. Thus, the ratio of soil strength to applied pressure, e.g., σ = (3.3c/ contact
pressure) is used to determine the thickness of the aggregate layer needed. Ex-
ample work sheets from the Mathcad 6.0 software used to make these calculations
follow.
26
EXAMPLE MATHCAD 6.0 WORKSHEET FOR
BARENBERG DESIGN METHOD
Solving of Boussinesq equation for z, the depth of aggregate for
wheel loads of 60 and 115 kN, applied contact pressure is 414 kPa.
With geotextile.
Wload = 115 000 newton ConPress = 414 000 Pa
r
Wload
ConPress
·
⋅ π
r = 0.29735 m
c = 10 000 Pa, 20 000 Pa … 60 000 Pa
σ c
6.0c
ConPress
( ) ·
Stuff in English units:
c
psi

r
ft
= 0.97557
1.45038
2.90075
4.35113
5.80151
7.25189
8.70226
z c,r r
1– c
1– 1– c
1
3
2
3
( ) ·
[ ]( )
( )
[ ]
σ
σ
z(c,r) c
0.89647 m 1.000 000 10
4
kg m
–1
sec
–2
0.58732 m 2.000 000 10
4
kg m
–1
sec
–2
0.43709 m 3.000 000 10
4
kg m
–1
sec
–2
0.33620 m 4.000 000 10
4
kg m
–1
sec
–2
0.25473 m 5.000 000 10
4
kg m
–1
sec
–2
0.17497 m 6.000 000 10
4
kg m
–1
sec
–2
This part of the sheet creates a database and makes a .prn
file out of it for importing to a spreadsheet later.
WRITEPRN (Baretwent) = z(c,r)
27
EXAMPLE MATHCAD 6.0 WORKSHEET FOR GIROUD AND
NOIRAY DESIGN METHOD
This is a file to calculate the aggregate depth needed for different K
values of the geotextile. It uses eq 43, 33, 35, 36, and 37 (a' > a, mean-
ing that the parabola between wheels is wider than the sum of the widths
of the parabolas under the wheels), 30 and 31 as well as 5 and 7 (on-
highway trucks). The original reference is Giroud and Noiray (1981).
K = 200 000 N m
–1
e = 2.0 m
tan α = 0.6 P = 230 000 newton
Pc = 414 000 P r = 0.3 m
Width of wheel load (on road), 5:
B
P
Pc
·
B = 0.745 m
h = 0, 0.1 … 1.0 m
Length of wheel load(on road), 7:
L
B
2
·
L = 0.527 m
Width of parabola under wheel, 30: Width of parabola between wheels, 31:
a(h) = 0.5 (B + 2 · h · tan α) aprime(h) = 0.5(e – B – 2 · h · tan α)
Settlement of geotextile from original position, 33:

s h
r aprime h
a h aprime h
( ) ·
⋅ ( )
( ) + ( )
Equation for half length of parabola under the wheel, 36:
b h a h 1 0.5 1
2 s h
a h
2
a h
2 s h
ln
2 s h
1
2 s h
a h
2
– 2 ( ) · ( ) + +
⋅ ( )
( )

]
]
]
]
+
( )
⋅ ( )

⋅ ( )
+ +
⋅ ( )
( )

]
]
]
]
¹
'
¹
¹
¹
¹
'
¹
¹
¹
|
.

`
,

]
]
]
]
]
]
Equation for half length of parabola between the wheels, 37:
The elongation of the geotextile, 35:
e h
b h bprime h
a h aprime h
–1 ( ) ·
( ) + ( )
( ) + ( )

]
]
]
bprime h aprime h 1 0.5 1
2 r – s h
aprime h
aprime h
2 r – s h
ln 2
r – s h
aprime h
2
( ) · ( ) ⋅ + +
⋅ ( ) [ ]
( )
¹
'
¹
¹
¹
¹
'
¹
¹
¹
|
.

`
,

+
( )
⋅ ( ) [ ]
⋅ ⋅
( )
( )
+
|
.

¹
'
¹
¹
¹
¹
¹
|
.

1
2 r – s h
aprime h
2
+ +
⋅ ( ) [ ]
( )
¹
'
¹
¹
¹
¹
'
¹
¹
¹
`
,

]
]
]
]
]
¹
'
¹
¹
¹
¹
¹
`
,

– 2
28
And the equation of thickness of aggregate vs. soil strength, 43:
5.181 10
4
kg m
–1
sec
–2
0 m
3.659 10
4
kg m
–1
sec
–2
0.1 m
2.722 10
4
kg m
–1
sec
–2
0.2 m
2.099 10
4
kg m
–1
sec
–2
0.3 m
1.663 10
4
kg m
–1
sec
–2
0.4 m
1.346 10
4
kg m
–1
sec
–2
0.5 m
1.111 10
4
kg m
–1
sec
–2
0.6 m
9.36 10
3
kg m
–1
sec
–2
0.7 m
8.058 10
3
kg m
–1
sec
–2
0.8 m
7.116 10
3
kg m
–1
sec
–2
0.9 m
6.477 10
3
kg m
–1
sec
–2
1 m
WRITEPRN(Gir20t) = Cu (K, P, h)
Cu (K, P, h) h
Cu K,P,h
1
2
P
2 B 2 h tan

K h
a h 1
a h
2 s h
2
( ) ·
+

⋅ + ⋅ ⋅ ( ) ⋅ + ⋅ ⋅ ( )
⋅ ( )
( ) ⋅ +
( )
⋅ ( )

]
]
]
¹
'
¹
¹
¹
¹
¹
¹
¹
¹
'
¹
¹
¹
¹
¹
¹
¹
π α α
ε
L h 2 tan
1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED
4. TITLE AND SUBTITLE 5. FUNDING NUMBERS
6. AUTHORS
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION
REPORT NUMBER
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING/MONITORING
AGENCY REPORT NUMBER
11. SUPPLEMENTARY NOTES
12a. DISTRIBUTION/AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE
13. ABSTRACT (Maximum 200 words)
14. SUBJECT TERMS 15. NUMBER OF PAGES
16. PRICE CODE
17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACT
OF REPORT OF THIS PAGE OF ABSTRACT
NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89)
Prescribed by ANSI Std. Z39-18
298-102
REPORT DOCUMENTATION PAGE
Form Approved
OMB No. 0704-0188
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestion for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington,
VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.
June 1999
Geotextile Reinforcement of Low-Bearing-Capacity Soils:
Comparison of Two Design Methods Applicable to Thawing Soils
Karen S. Henry
U.S. Army Cold Regions Research and Engineering Laboratory
72 Lyme Road
Hanover, New Hampshire 03755
Special Report 99-7
Approved for public release; distribution is unlimited.
Available from NTIS, Springfield, Virginia 22161
Office of the Chief of Engineers
Washington, DC 20314-1000
36
UNCLASSIFIED UNCLASSIFIED UNCLASSIFIED UL
Geosynthetics Military vehicles
Geotextiles Reinforcement
Low-bearing-capacity soils Thawing soils
WU: TO-007
4A 161102AT4Z
Thawing fine-grained soils are often saturated and have extremely low bearing capacity. Geosynthetics are used to
reinforce unsurfaced roads on weak, saturated soils and therefore are good candidates for use in stabilization of
thawing soils. To stabilize the soil, a geotextile is placed on it, then the geotextile is covered with aggregate. Design
involves selection of aggregate thickness and geotextile. There are two commonly used design techniques for
geotextile reinforcement of low-volume roads, and the Army uses one of them. The theory and use of the two
design methods for static loading (i.e., up to 100 vehicle passes) are presented and compared in this report. The
design method not used by the Army offers the potential to reduce aggregate thickness over the geotextile because
it accounts for the fact that the geotextile helps support the traffic load (when in tension) and confines the soil
between the wheels and the subgrade. However, this alternative method appears to be unconservative with re-
spect to stresses estimated at the subgrade surface. Thus, the current Army design technique should be used until
more research is conducted. In the meantime, straightforward design curves for Army 10- and 20-ton trucks as
well as vehicle loading and tire pressure information for a number of other vehicles are included in this report to
help make the current design method easy to use.
Future work should consider adopting a hybrid design method that provides realistic estimates of stresses at the
subgrade and accounts for the tensile properties of geotextiles. In addition, aggregates other than the high-quality
crushed rock that is inherently assumed by each design method should be accounted for in new design develop-
ment.

99-7

US Army Corps of Engineers®
Cold Regions Research & Engineering Laboratory

Geotextile Reinforcement of Low-Bearing-Capacity Soils
Comparison of Two Design Methods Applicable to Thawing Soils
Karen S. Henry June 1999

SPECIAL REPORT

Abstract: Thawing fine-grained soils are often saturated and have extremely low bearing capacity. Geosynthetics are used to reinforce unsurfaced roads on weak, saturated soils and therefore are good candidates for use in stabilization of thawing soils. To stabilize the soil, a geotextile is placed on it, then the geotextile is covered with aggregate. Design involves selection of aggregate thickness and geotextile. There are two commonly used design techniques for geotextile reinforcement of lowvolume roads, and the Army uses one of them. The theory and use of the two design methods for static loading (i.e., up to 100 vehicle passes) are presented and compared in this report. The design method not used by the Army offers the potential to reduce aggregate thickness over the geotextile because it accounts for the fact that the geotextile helps support the traffic load (when in tension) and confines the soil between the

wheels and the subgrade. However, this alternative method appears to be unconservative with respect to stresses estimated at the subgrade surface. Thus, the current Army design technique should be used until more research is conducted. In the meantime, straightforward design curves for Army 10- and 20-ton trucks as well as vehicle loading and tire pressure information for a number of other vehicles are included in this report to help make the current design method easy to use. Future work should consider adopting a hybrid design method that provides realistic estimates of stresses at the subgrade and accounts for the tensile properties of geotextiles. In addition, aggregates other than the highquality crushed rock that is inherently assumed by each design method should be accounted for in new design development.

How to get copies of CRREL technical publications: Department of Defense personnel and contractors may order reports through the Defense Technical Information Center: DTIC-BR SUITE 0944 8725 JOHN J KINGMAN RD FT BELVOIR VA 22060-6218 Telephone 1 800 225 3842 E-mail help@dtic.mil msorders@dtic.mil WWW http://www.dtic.dla.mil/ All others may order reports through the National Technical Information Service: NTIS 5285 PORT ROYAL RD SPRINGFIELD VA 22161 Telephone 1 800 553 6847 or 1 703 605 6000 1 703 487 4639 (TDD for the hearing-impaired) E-mail orders@ntis.fedworld.gov WWW http://www.ntis.gov A complete list of all CRREL technical publications is available from: USACRREL (CEERD-IM-HL) 72 LYME RD HANOVER NH 03755-1290 Telephone 1 603 646 4338 E-mail techpubs@crrel.usace.army.mil For information on all aspects of the Cold Regions Research and Engineering Laboratory, visit our World Wide Web site: http://www.crrel.usace.army.mil

distribution is unlimited. Henry June 1999 Prepared for OFFICE OF THE CHIEF OF ENGINEERS Approved for public release.Special Report 99-7 US Army Corps of Engineers® Cold Regions Research & Engineering Laboratory Geotextile Reinforcement of Low-Bearing-Capacity Soils Comparison of Two Design Methods Applicable to Thawing Soils Karen S. .

ii .PREFACE This report was written by Dr.S. Citation of brand names does not constitute an official endorsement or approval of the use of such commercial products. The contents of this report are not to be used for advertising or promotional purposes. The author thanks Professor Mark Evans of West Point Military Academy for technical review of the manuscript. Hanover. Research and Engineering Directorate.S. Army Corps of Engineers Cold Regions Research and Engineering Laboratory. TO-007. His review questions required responses that added significantly to the breadth and depth of this work. She also thanks George Blaisdell of CRREL for technical review as well as for numerous discussions of this topic and encouragement. Holtz of the University of Washington are acknowledged. Research Civil Engineer of the Civil Engineering Research Division. Funding for this work was provided by the U. AT42. Karen Henry. U. New Hampshire. This work was initiated as an independent study project while the author was a graduate student under his guidance. Use of Geosynthetics to Rapidly Reinforce Thawing Soils. Army Corps of Engineers under Project/Contract No. Finally. the contributions of Professor Robert D.

...................................................... resulting from the use of geotextile as a function of soil cohesion for 12-in......... and equivalent contact area used in analysis .............................................................. ho’................................................................................. 11 11.................................................. Comparison of Giroud and Noiray method with Army method ..... ho’........ Applicability for use by the U.................................................. and shape of deformed geotextile . ∆h........................................................................................................................... ii 1 1 3 4 6 6 6 9 12 13 13 13 15 15 21 21 23 25 29 ILLUSTRATIONS Figure Page 1.............................................................................................. Aggregate thickness.................... 8 8................................................. Introduction ....................... rut depth ..............................CONTENTS Page Preface ................... and Giroud and Noiray ..... 2 3.................... Validation of calculation techniques ...... Stress distribution through the aggregate layer .................... Wheel load distribution by the aggregate layer to the subgrade ......................................... Diagram of “membrane effect” of geotextile reinforcement of thawing soil..... Theory .......................................... Aggregate thickness design curve for dual-wheel load on gravelsurface roads ............................................................................. and design points generated for this report according to method documented in Appendix B ........... Alternative method presented by Giroud and Noiray .... Design aggregate thickness chart for 88....................................................................................... 4 5......................... ∆h............................................ Wheel and axle configurations ........................... Abstract ........................ Design example ... 10 10........ Definition of tire contact area for dual tires....... Army use on thawing soils ..... Appendix A: Geotextile survivability requirements .........................................S... 7 7................... Static loading design curves from Barenberg et al............................................................................. and reduction of aggregate thickness............................. Army ............................................................................. and reduction of aggregate thickness...................................... 3 4.......................................................... 9 9............................ Literature cited ............................................................................. Design example .................................. Aggregate thickness design curve for single-wheel load on gravelsurface roads ...................................... resulting from the use of geotextile as a function of soil cohesion for 8-in........................................... Aggregate thickness design curve for tandem-wheel load on gravelsurface roads ............ Applicability for U........................ Aggregate thickness.......................... 14 iii ......................................................................95-kN wheel load ..........................S........................ 5 6................................................................................................................................................................................ Recommendations for future work ................................. Summary and conclusions ........... 2 2............. Design curves for Army vehicles ................................................................................................................................................................... Method currently used by the Army .......... Appendix B: Methods used to recalculate design curves from Barenberg et al........... Theory ................................................................................................................. rut depth ......................

14 15 16 17 18 19 TABLES Table Page 1...................................and 20-ton trucks with tire pressure of 414 kPa ......... Design curves for static loading for 10-ton dump truck..... according to Barenberg method and Giroud and Noiray method ................. 15. Static loading design curves without geotextile from Giroud and Noiray and points generated for this report according to method documented in Appendix B ..................15 m ............... Maximum vertical stress at various depths below applied wheel load of 115 kN and contact pressure of 414 kPa according to Newmark and trapezoidal stress distribution used by Giroud and Noiray ....... Static loading design curves adapted from Giroud and Noiray and Barenberg et al........................................ 14...... theoretical prediction by Giroud and Noiray for unreinforced test sections and reinforced test sections .. 16..............................................12........ Traffic loading data for Army vehicles .......................... 6 2................................................................... Thickness of plastic zone in the subgrade for dual-tired truck loading and aggregate layer of thickness 0................................................................ according to Barenberg method and Giroud and Noiray method ....... 13................... Field performers vs........... 17.... 16 iv ..... for 10...... Static loading design curves with geotextiles from Giroud and Noiray and points generated for this report according to method documented in Appendix B ................ Tensile modulus values of geotextiles at 5% strain and at failure based on information in Geotechnical Fabrics Report ................................................................ 11 3. 13 4........... Design curves for static loading for 20-ton dump truck.......

* The design method is presented in TM5-818-8 (1995) as a series of soil strength vs. Geotextiles have been used in the construction of low-volume. Although both design methods include traffic loading for up to 1000 vehicle passes.. with a tire pressure of 552 kPa (80 psi) (Fig. Information about Army vehicle loading and design curves for specific Army vehicles is provided in this report to help make the design technique easier to use. and determining the depth of aggregate required with and without a geotextile. aggregate thickness design curves for various wheel loads (defined below). Although TM5-818-8 does not specify the aggregate properties required for low-volume roads.0 with geotextile and 3. which involves a maximum of 100 vehicle passes. Nc.Geotextile Reinforcement of Low-Bearing-Capacity Soils Comparison of Two Design Methods Applicable to Thawing Soils KAREN S. 4) Using the product cNc in the appropriate design chart (e. c. Army for the stabilization of low-bearing-capacity soils for low-volume roads and trails with geotextile and aggregate was developed by the U.).g. The design procedure includes 1) Converting soil strength to an equivalent cohesion. the geotextile is placed directly on the soil and then covered with aggregate. Nc values used with design curves for static loading are 6. Static loading is defined as up to 100 passes of a vehicle at the maximum wheel load and a minimum rut depth of 0. 2) Selecting a maximum wheel load. cohesive soils by Barenberg et al.3 without geotextile. (1975). Forest Service (Steward et al. here we deal only with design for static loading. 6) If use of a geotextile is advantageous. specifying one according to geotextile construction survivability requirements.S. HENRY INTRODUCTION Thawing fine-grained soils are often saturated or even supersaturated and thus have extremely low bearing capacity. The current Army design technique for static loading of low-volume roads on low-bearing-capacity soils was examined for ease of use and applicability to the reinforcement of thawing soil. The design involves selecting aggregate thickness and the geotextile. .S. METHOD CURRENTLY USED BY THE ARMY (TM 5-818-8) The design method currently used by the U. 3) Selecting a value for a bearing capacity factor. (1975). There are two commonly used techniques for designing for soil reinforcement using geotextiles. 1 through 3).S. To stabilize weak soil with a geotextile for trafficking. 1). unsurfaced roads on weak and saturated soils to reinforce the base course–subgrade interface and therefore are good candidates for use in stabilization of thawing soils. 5) Determining which section is less costly to build. Theory and results from both design methods are presented in this report. Another design method that offers the potential to reduce required aggregate thickness over the geotextile (and thus cost) was compared with the Army method.10 m (4 in. one of which is prescribed in U. Army guidance. *The low-bearing-capacity soils were assumed to be soft. Fig. specifically for Army vehicles. 1977) based on theory presented by Barenberg et al.

.000 lb) 53. Soils that yield these values include crushed rock. (From TM 5-818-8. thus.000 lb) 44.2 kN ( 5.5 Wheel Load. it is now provided.2 kN (16.0 P = 2W 10 1 5 50 10 cNc 100 50 500 kPa 100 psi Figure 1.0 kN ( 9. (From TM 5-818-8.in.5 1. and well-graded gravel (e. Aggregate thickness design curve for dual-wheel load on gravel-surface roads.8 kN ( 4. 70 60 m 1.000 lb) 66. 1975).) aggregate that meets base course requirements is presumably required.5 kN (10.000 lb) 0. FM5-430-001.0 kN (20.5 1.000 lb) 22. W 1. 70 60 Aggregate Thickness 50 40 30 20 10 0 m Wheel Load. For single and in. mechanically stabilized aggregates.7 kN (15. Holtz and Kovacs 1981).000 lb) 17.0 kN (20.5 P = 2W 0 10 1 5 50 10 cNc 100 50 500 kPa 100 psi Figure 2.000 lb) Aggregate Thickness 50 40 30 20 10 0 0 0.000 lb) 40. Field Manual FM5-430-001 (1994) recommends that base course material have minimum CBR values of 80 to 100.4 kN (12. Aggregate thickness design curve for single-wheel load on gravel-surface roads.) 2 . W 89.0 89.000 lb) 71. Further evidence for the requirement of high-quality aggregate is the fact that the design technique is based on experiments that used crushed-rock aggregate (Barenberg et al.g. Guidance for selecting wheel loads and contact pressures to use with this design is not given in TM5-818-8.

Calculate cNc as (52)(3.5 x 1.) dual wheels on a single axle (Fig. is defined as the total load on the axle. The axle load.5 psi) or CBR of 2. determine the aggregate thickness required with and without geotextile for a soil cohesion of 52 kPa (7. Contact pressures for use in design are approximately 0.g. P. Barenberg et al. 3 .66 times the total load on one side of both of the tandem axles.32 m (13 in.5 50 40 30 20 10 0 0 0. 1975).0 P= 2W a Wheel Load. For tandem axles (Fig.) of aggregate is required with geotextile. 4c and 4d).12 m (5 in. or 2W.3) = 172 kPa (25 psi) without geotextile..563 times the maximum allowable single axle load (Yoder and Witczak 1975).000-lb) maximum expected single axle load (40-kN or 9. contact pressure (e. Determine whether the cost of the geotextile exceeds the cost of 0.6 kN (43. a quantity used in other design methods. is defined as the total load on either the left or right side of the axle.000 lb) a = 0.† There will be approximately 100 passes of this vehicle. Fig. and a rut depth of 0. 5).0 times the tire inflation pressure for single-tired vehicles and 0. Aggregate thickness design curve for tandem (dual)-wheel load on gravel-surface roads. 4.500 lb) 106.) can be tolerated. Most U. 1. there is negligible difference in the aggregate thickness design curves for actual tire inflation pressure vs. Enter Figure 2 with a cNc value of 172 kPa (25 psi) for a wheel load of 40 kN (9.60. 0.6 kN ( 8. estimations of wheel and axle loads vary.8 kN (24. (From TM 5-818-8. W.9 to 1.750 lb) 166.000 lb) 77. However. 3.20 m (8 in. Design example Given an 80-kN (18.000-lb wheel load) on a dual-tired vehicle.S. If it is advantageous to use a geotextile. The tire inflation/contact pressure is equal to 552 kPa (80 psi).500 lb) 35. 1975) *W = Sum of load on all four wheels on one side of the axle 10 1 5 50 10 cNc 100 50 500 kPa 100 psi Figure 3. for the design method presented in TM5-818-8. 70 60 Aggregate Thickness m 1. †Saturated silts and clays are likely to have CBR values of this order of magnitude. Using a cNc value of 312 kPa (45 psi) for the same wheel load.in.) aggregate. specify one using Tables 2-2 through 2-4 in TM5-818-8 (based on the need for the geotextile to survive *The relationship between shear strength (cohesion) or CBR and Cone Index is given in Figure 2-3 of TM5-8188 (1995). Thawing frost-susceptible soils may have even lower CBR values.8 kN (37. 2.8 kN (17. states allow maximum loads on each pair of tandem axles equal to 0. Giroud and Noiray (1981) obtained the design axle load by multiplying the sum of the two axle loads by 0.000 lb) to obtain a value of 0.66 (Barenberg et al. 4a and 4b).*.70 to 0.3 m (12 in. (1975) use a wheel load of 0. and (52)(6) = 312 kPa (45 psi) with geotextile. the wheel load. which would be saved by using the geotextile. W* 194.) of aggregate required without geotextile.75 times the tire inflation pressure for dual tires (Barenberg et al.

0 m (d) Single Wheels. p is the effective stress. Wheel and axle configurations. qd.(a) Single Wheel. the angle of internal friction is zero and the undrained shear strength of the soil is equal to its cohesion. the guidance presented in Appendix A. moving wheel loads are transient. Single Axle e l = 1. For soft clay subgrades at or near saturation. Thus. Tandem Axles e Figure 4. meaning that undrained loading applies.8 to 2. Tandem Axles e l = 1. (2) . Single Axle e (b) Dual Wheels. the ultimate bearing capacity. construction). Alternatively. which was developed more recently than that listed in TM5-818-8 (1995). can be used to select a geotextile. Theory Bender and Barenberg (1978) summarize the theory and tests that led to the design method described above.0 m (c) Dual Wheels. and φ is the angle + t of internal friction.8 to 2. Based on the theory of plastic equilibrium. the shear strength is s = c + p tan φ (1) 4 where s is the shear strength of the soil. Using the Mohr–Coulomb failure criteria for soils. for soil in this condition is qd = (2 + π)c . c is the cohesion.

without geotextile. Barenberg et al.) However. In addition to the change in failure mode from local to general bearing capacity failure. (1975) did not consider tensile modulus or strength (or any mechanical property) of the geotextile in developing their design method. (3) Barenberg et al.. Barenberg et al.8 0. this relationship was σzallowable = 3. 1975. cyclic loading) with a geotextile (Mirafi 140) placed between crushedrock aggregate and a saturated clay subgrade.25 1 With Geotextile Contact Pressure = 55.000-lb) wheel load. (1975) conducted laboratory tests (two dimensional. A.0 0. soil strength by assuming that the allowable pressure at the subgrade is 3. The radius.3c without geotextile and 6c with geotextile (Fig.00 30 5 1.2 40 Aggregate Thickness 1. even though Bender and Barenberg (1978) note that “a layer of aggregate material is always needed on top of the fabric to anchor it so that the necessary tensile forces can be developed in the fabric” (p. Stress levels on the subgrade were estimated by using a Boussinesq stress distribution beneath a circularly loaded area (e.75 4 1. (From Barenberg et al. Furthermore.4 kPa (150 psi) 10 2 0. 5 .50 20 3 0.in. 5).3c . The unreinforced systems deformed progressively with repeated loading. 66). was determined by dividing the wheel load by the contact pressure. needed for determination of the stress at the subgrade surface. The effects of traffic loading when vehicle passes exceed 100 were accounted for by Steward et al.25 40 6 1. 50 m 1. (4) However. r.50 7 50 kPa psi 0 Undrained Shear Strength Figure 5. Design aggregate thickness chart for a 88.g. The allowable stress with geotextile on the subgrade surface was σzallowable = 6c .2 kPa ( 80 psi) Contact Pressure = 103. was obtained from A = πr2. The contact area. localized plastic strains that can cause localized shear failure begin at q ≈ πc . (1975) or Bender and Barenberg (1978).2 0 0 0 CBR 0 0. Stress at the subgrade was calculated by using Boussinesq stress distribution beneath a circularly loaded area (Newmark 1942). (1975) constructed design charts for aggregate thickness vs.95-kN (20. and ratios between the calculated subgrade stress and measured soil strength were developed.6 30 Without Geotextile 20 0. neither the minimum depth for anchorage nor the mechanical properties of the aggregate layer are specified in either Barenberg et al.4 10 0. (5) These numbers are very close to the theoretical values of general and local bearing capacity failure (eq 2 and 3). Newmark 1942). the soil systems that contained geotextile reached a level of permanent deformation so that further loading of the same magnitude caused negligible additional deformation. (1977) by reducing the Nc values.

680 77. 1 through 3) apply to vehicles with a 552 kPa (80 psi) tire pressure. this method was not intended for shallow thaw layers. CRREL. Vehicle/axle type HMMWV M939 (6 × 6) 5-ton cargo truck HEMTT.31/7. determine the aggregate thickness required with a geotextile of modulus K of 100 kN/m (570 lb/in..48/68. Conditions: 80 kN load on a dualwheel single axle. Thus. The Giroud and Noiray (1981) design method was chosen for comparison with the method now used by the U.59/31. ALTERNATIVE METHOD PRESENTED BY GIROUD AND NOIRAY Giroud and Noiray (1981) developed a design method for geotextile placement between the aggregate and subgrade of unpaved roads based on bearing capacity theory for static loading.5 psi). The maximum wheel load for tandem axles listed in Table 1 was determined by multiplying the total load on the rear axles by 0. Since TM5-818-8 does not list typical wheel loads for Army vehicles (nor does U.g. some are provided for vehicles most likely to be used on low-volume roads or trails (Table 1).520 Tire pressure (kPa/psi) 241/35 345/50† 483/70† Not available 414/60 414/60 Ground clearance (m/in.94*/4. (1975) demonstrated negligible difference in the design curves due to variation in the contact pressures ranging from 552 to 1034 kPa (80 to 150 psi).).S.24 to 88. 1997). Army because it is widely used (e. soil cohesion of 52 kPa (7.000 lb). M985 M125 10-ton truck Articulated 8 × 8 M917 20-ton dump truck Gross vehicle weight (kN/lb) 33.S.41/16 0.83/2.400 113.30/12 0. Bounds on the depth of the thawed layer for full development of the plastic zone are discussed in the following section.80/33. approximately 100 passes of the vehicle. The curves are used to determine aggregate thickness without geotextile (ho′) and reduction of aggregate thickness with geotextile (∆h) for geotextiles with different values of tensile modulus. with an axle load of 80 kN (18. K. the depth of which depends on the geometry and magnitude of the loading.906 Load on rear axles (kN/lb) 21.000 324.4/12.0/38.000) at various rut depths (Fig. The method accounts for the load support and soil confinement provided by the geotextile itself and is presented as a set of curves for dual wheels on a single axle.000 302.906 Maximum wheel load (kN/lb) 10.346 50..58/9.000 Not available 324. little difference in aggregate thickness is Table 1. it was assumed that 65% of the gross vehicle load is applied on the rear axles. 1995) and because of its potential for cost savings by allowing thinner aggregate layers over the geotextile because it takes into account the tensile properties of the geotextile. Although Figure 6 is constructed for a tire inflation pressure of 480 kPa (70 psi).3 m (12 in.434 41. Army Field Manual FM5-430-00-1). tire inflation pressure of 552 kPa (80 psi).30/72.000 187. Giroud and Noiray 1981) then dividing by two. Holtz et al.) and without a geotextile.14/65. and it would be conservative to use this method for shallow thawed layers.225 258. The maximum wheel loads listed in Table 1 for Army vehicles are reasonably well-represented in Figures 1 through 3.30/72. For the 20-ton dump truck (three rear axles). †From Jeffrey Stark (personal communication.96 kN (5.g.7/11.51/25. Traffic loading data for Army vehicles (Foss 1983).52/20 0.30/12 0.400 56.4/17.60 (e.000 to 20. Design example Given the same vehicle and soil conditions described in the previous design example for the method currently used by the Army.869 138. Barenberg et al.0/58. for single and dual wheel loads varying from 22. 6 . This theory and design method assume that the subgrade is uniform and that full plastic failure zones can develop. and a tolerable rut depth of 0.489 146. Even though the tire pressures in the figures are higher than the tire pressures listed in Table 1.156 169.70 then dividing it by two.000 289.Applicability for Army use on thawing soils The design curves supplied in TM5-818-8 (Fig.65*/4.30/12 Not available Notes: *No rear axle load was given.) 0. the maximum wheel load was estimated by multiplying the total load on the rear axles by 0. 6 and 7).

However.3 m (12 in.) of aggregate required without a geotextile. resulting from use of geotextile as a function of soil cohesion for 12-in. ho′ (top). Barenberg et al. (From Giroud and Noiray 1981.0 Axle Load = 80 kN (18. m 1. Enter the top chart in Figure 6 with a soil cohesion value.6 0. 1. rut depth.12 m (5 in.000 lb) Rut Depth = 0. 2.6 2 0. Giroud and Noiray 1981).06 m (3 in.3 m (12 in. the 7 . N is number of vehicle passes and K is tensile modulus of geotextile.2 5 0 0 0 UndrainedShear Strength 0 CBR 0 in.) Tire Pressure = 480 kPa (70 psi) 35 Aggregate Thickness. ∆h 35 30 25 20 15 10 0.in.20 m (8 in. the aggregate thickness theoretically allowed when a geotextile is used is 0.) Tire Pressure = 480 kPa (70 psi) 0. and for a geotextile with K = 100 kN/m and determine that the reduction in aggregate thickness ∆h allowed is 0. Thus. for 100 vehicle passes. and reduction of aggregate thickness. Aggregate thickness. of 52 kPa.g. for 100 vehicle passes and determine that ho′ is 0.). c.4 3 1 2 3 4 5 6 K = 450 kN/m K = 400 kN/m K = 300 kN/m K = 200 kN/m K = 100 kN/m K = 10 kN/m 80 100 16 120 18 140 kPa Undrained Shear Strength 0 0 1 Figure 6.0 1 2 20 4 40 6 N = 100 N = 10 60 8 2 10 80 12 3 100 14 16 4 120 kPa psi KH-78 Decrease in Aggregate Thickness. ho 30 25 20 15 10 0..2 5 0 0 0 20 40 60 4 5 6 0. m 1.8 1 Axle Load = 80 kN (18. ∆h (bottom). 1975.4 0.000 lb) Rut Depth = 0.) expected (e.8 0.). Enter the bottom chart in Figure 6 with a soil cohesion value of 52 kPa.

) Tire Pressure = 480 kPa (70 psi) 35 Aggregate Thickness. below.) aggregate layer should probably be a minimum of 0.4 0.). ho 30 25 20 15 10 0. less aggregate is recommended by the Giroud and Noiray design method than by the Army method—0.in.6 0. this is discussed in Theory.10 to 0.) to prevent damage to the geotextile as discussed below.2 m (8 in. ho′ (top).2 5 0 0 0 20 2 4 1 40 6 8 2 60 10 N = 100 N = 10 80 12 3 100 14 16 120 kPa psi 4 Undrained Shear Strength 0 CBR 0 in. approximately the same aggregate savings are realized by .2 m (8 in. Notice that for the same conditions without geotextile. Although the total amount of aggregate required by the two design methods is different. and reduction of aggregate thickness.6 3 1 2 20 40 60 80 100 16 120 18 140 kPa Undrained Shear Strength 0 0 Figure 7. Aggregate thickness.) compared with 0.2 m (8 in.000 lb) Rut Depth = 0.2 5 0 0 0 0. ∆h 35 30 25 20 15 10 0. The difference results from how each method estimates stress at the subgrade.8 ∆h for: 1 K = 50 kN/m 2 K = 200 kN/m 3 K = 350 kN/m Axle Load = 80 kN (18.8 0.000 lb) Rut Depth = 0. resulting from use of geotextile as a function of soil cohesion for 8-in.0 Decrease in Aggregate Thickness.0 Axle Load = 80 kN (18.15 m (4-6 in. (From Holtz and Sivakugan 1987.4 0. m 1. m 1. ∆h (bottom). and in Stress Distribution Through the Aggregate Layer in the comparison of the Giroud and Noiray and Army methods. rut depth. N is number of vehicle passes and K is tensile modulus of geotextile.25 8 m (10 in.) Tire Pressure = 480 kPa (70 psi) 0.

(1975). and shape of deformed geotextile (bottom). If a geotextile with a modulus of 300 kN/m (1. or essentially 5 in.) aggregate cover over a geotextile is recommended at all times. It also helps anchor the geotextile to allow it to develop the required tension. of the geotextile is increasingly influential as the rut depths increase (membrane action occurs at large strains).15 m (4 to 6 in. The aggregate protects the geotextile from damage imposed by construction traffic as well as degradation due to exposure to ultraviolet light (sunlight).). Theory Giroud and Noiray (1981) assumed a soft. making it stronger (Fig. saturated cohesive subgrade in undrained loading. (1975).13 m.using the Giroud and Noiray method with a 100 kN/m modulus geotextile and the Army method (0.) 9 .) is used. however.10 to 0. K. The tensile strength of the geotextile then helps to both support the load and confine the soil above the geotextile. see e..g. 8). for each case).2 m (8 in. (After Giroud and Noiray 1981. Thus. Whitman and Hoeg 1965) to general (plastic). However. The modulus. the geotextile undergoes deformation that puts it in tension.715 lb/in. and that the effect of the geotextile placed between the aggregate and the subgrade will change the bear- ing capacity failure from local (near the elastic limit. this theory and the design technique is based on the assump- Aggregate Geotextile Subgrade Soil e 2a β t t β 2a (P ) β t 2a β t A (P) Initial Location of Geotextile s B r B s A e Geotextile Figure 8. As bearing capacity failure deforms the subgrade. 0. Membrane effect refers to the fact that the material contained by the concave side of a stretched. a minimum 0. Figure 6 indicates that no aggregate is required since ∆h = 0. However. they applied the same soil mechanics principles as did Barenberg et al. Giroud and Noiray (1981) extended this concept to account for the “membrane effect” of the geotextile.12 m vs. Like that of Barenberg et al. Diagram of “membrane effect” of geotextile reinforcement of thawing soil (top). flexible membrane is at a higher pressure that the material on the outside of it.

The aggregate will not slide along the geotextile surface. 3. If a geotextile is used and the full plastic zones do not develop. and the same is assumed for the current U. (7) Table 2 presents values for Hmin for a minimal ag- e (a) Aggregate Ac (b) Soil mechanically associated with adjacent tires Figure 9. Army design technique. Giroud and Noiray (1981) assumed the value of tan α to be 0. and. the geotextile may provide important separation between the thawing soil and the aggregate that will likely lead to longer use of the road without maintenance. However. and 5. Assuming a dual-tired truck with an axle load. The shape of the deformed geotextile consists of parabolas (Fig. K. 8. Pc. and equivalent contact area used in analysis (Giroud and Noiray 1981) (bottom). is uniform along the entire length of the geotextile. having tire pressures. Hmin.tion that the subgrade soil is of a sufficient depth.6. the width. h is the thickness of the aggregate layer.) thick. L and B refer to length and width of single tire print. designing for reinforcement with geotextile would be conservative. 9b). the tension in the geotextile will not be fully mobilized. assuming that the subgrade soil underlying the thawed soil is stronger than the thawing soil. where φ is the friction angle of the base course expressed as radians (Fig. and α = (π/4) – (φ/2). When the thawed layer is so thin that full plastic zones cannot develop. The elongation. For the stress distribution assumed by Giroud and Noiray (1981). the Giroud and Noiray (1981) technique is not generally applicable to thawed (or weak) layers less than 0. Definition of tire contact area for dual tires (top). 2. of the wheel load is given by (Giroud and Noiray 1981) B= P Pc gregate cover of 0. P. (For single tires.4 m (16 in. 10). or strain. to allow the plastic zones associated with ultimate bearing capacity to develop. 4. B (m). Thus. respectively. The geotextile does not fail. used in de- for off-highway trucks and B= P 2 Pc for on-highway trucks.S.) L L B B 10 . the support required of the geotextile will also be less than if the subgrade were uniformly weak. The modulus of the geotextile. Other assumptions pertaining to the geosynthetic include that 1.15 m (6 in.). bottom). therefore (as mentioned above). this amounts to H min = B + 2 h tan α 2 (6) where B is the width of the loaded area at the soil surface (Fig.

and 7. and 5. Fannin and Sigurdsson 1996). The influence of the inaccuracy of this assumption on the validity of the design technique is not yet clear.42/16. use of a high-quality aggregate (as assumed by this method) would probably prevent this from occurring. However.1 to 1.3. however. Giroud and Noiray (1981) recommend the use of a biaxial tensile test. Kinney 1982).g. Axle load kN (lb)/ tire pressure kPa (psi) 80 kN (18.2%. 1998. 3. Kinney and Barenberg 1979. Thickness of plastic zone in the subgrade for dual-tired truck loading and aggregate layer of thickness 0. while the maximum strain in each of the geotextiles was 2.000 lb)/ 480 kPa (70 psi) 60 kN (13. 2. where the lateral deformation of the geotextile is prevented during testing. For example.. Fannin and Sigurdsson (1996) measured the strain in three geotextiles placed beneath a 0.47/18. Measurements of test sections that were carefully trafficked to minimize the wander of wheels in lanes indicated that the shape of the deformed geotextile is approximately parabolic (e. The strain in the three geotextiles at 80 passes of a standard axle load of 80 kN and 620 kPa tire pressure averaged 0. Wheel load distribution by aggregate layer to subgrade (Giroud and Noiray 1981).6.1.). Tests have also shown that the strain is not completely uniform along the length of the geotextile (e.. offhighway truck Hmin (m/in.5 Plastic zone thickness.g. Kinney and Barenberg 1979. 0.. Fannin and Sigurdsson 1996).25-m-thick layer of sandy gravel and over a subgrade of average strength of 40kPa. Soderman and Giroud 1995).5 *The tensile modules of geotextiles can now be obtained from ASTM D 4595 (1998) Standard test method for tensile properties of geotextiles by the wide-width strip method.38/15 0.43/17 11 . respectively. personal comTable 2. Kinney (1982. the aggregate can (and most likely does) slide along the surface of the geotextile (e.) 0. and that the secant modulus in the transverse direction of the road be used.500 lb) 480 kPa (70 psi) Plastic zone thickness. It is also not clear what modulus value should be used with this design method.9.B L Aggregate Surface α γh γh h 2a Geotextile Subgrade/Aggregate Interface Figure 10.g.15 m (6 in.4%. sign is the secant modulus obtained from tensile tests.35 times (e.g.) 0. onhighway truck Hmin (m/in.* The first assumption is reasonable for geotextiles that meet survivability requirements. Biaxial testing would lead to estimates of modulus values that are higher than those determined in uniaxial tests by about 1. If the geotextile–aggregate friction is inadequate. Giroud 1992. Kinney and Barenberg 1979..

Giroud and Noiray (1981) assumed a single-axle dual-wheel configuration. there are two more ways in which the Giroud and Noiray theory differs from that presented by Barenberg et al. (1975). design curves for Army vehicles should be developed for higher axle/wheel loads and for variations in the shape of the applied loading. 9). However. of some products commercially available today are significantly greater than those for which Giroud and Noiray (1981) provided design curves. Table 3 indicates that the tensile modulus values in the cross-machine direction. are presented in Table 3. Thus. Therefore.munication) claimed that repeated loading of the only geotextile that he tested resulted in an effective modulus that was “many times lower” than that determined from monotonically loaded uniaxial and biaxial tests. Fannin and Sigurdsson 1996).S. (1975) did not discuss the mechanical properties of aggregate. although the tests they performed utilized crushed-rock aggregate.. Estimated axle loads for U. Based on limited field experiments. in typical stress–strain relations for tensile loading of needlepunched geotextiles. Army vehicles range up to 324 kN (73. Applicability for use by the Army The design curves based on the Giroud and Noiray (1981) method currently published are for standard 80-kN (18. 10) as opposed to the Boussinesq distribution beneath a circular plate used by Barenberg et al. Most significant is the shape of the stress distribution through the aggregate layer to the subgrade. 5% strain appears to be a reasonable estimate for static loading of geotextiles performing reinforcement over low-bearing-capacity soils (e. Fig. Geotextile modulus values at 5% strain.. field or other experimental work is needed to help establish the effective modulus values of B 2 B 2 for off-highway trucks and for on-highway trucks (9) where L is the length of the rectangle formed by a set of dual wheels (Fig. but Barenberg et al. This suggests that the design method should include higher modulus values. c) 2a’ = e – B – 2h tan α (11) (10) 12 . The difference is especially significant for relatively thin aggregate layers (less than approximately 0. (1977). provided by the manufacturers for a variety of geotextiles. the design equations from Giroud and Noiray are a) L = L= where e and a′ are defined in Figure 8 (bottom). but possibly far lower for repeated loading than for monotonically loaded uniaxial tests. research to determine the effective modulus values when geotextiles are being repeatedly loaded or trafficked in-situ would be useful.000-lb) axle loads for on-highway trucks with tire inflation pressures of 480 and 620 kPa (70 and 90 psi) (e. 6 and 7).g. the HEMTT). In reinforcing low-bearingcapacity soil.. as will be demonstrated in the next section. Therefore. The method published by Giroud and Noiray (1981) uses geotextile tensile modulus values ranging from 10 to 450 kN/m. Giroud and Noiray (1981) used a trapezoidal distribution of the stress beneath a loaded rectangle (Fig. The assumed shape of the load and the assumed stress distribution through the aggregate layer to the subgrade results in significant differences in the estimated stresses at the subgrade for certain loading and soil conditions. the direction that would be transverse to traffic. d) s = ra ′ for a > a´ and a + a′ s= 2ra 2 2 a 2 + 3 aa ′ – a ′ 2 for a > a´ (12) where s and r are defined in Figure 8 (bottom). whereas many Army vehicles have single tires on tandem axles (e. Giroud and Noiray (1981) also assumed a minimum CBR value of 80 for the overlying aggregate.g.3 m or 12 in. Furthermore. In addition to including the tensile support provided by the geotextile. In addition.). and tire pressures can be as low as 241 kPa (35 psi) (Table 1). In addition to eq 8.g.000 lb). consideration should be given to the shape of the wheel load applied on the surface. the slope of the stress–strain curve is initially quite low. if this design technique were to be adopted by the Army. resulting in low modulus values at low strains. recall from the above discussion that modulus values are higher in biaxial tension. the geotextile is expected to undergo both biaxial tension and repeated loading. b) 2a = B + 2h tan α where a is defined in Figure 8b.

(1975) and Giroud and Noiray (1981) for static loading were reconstructed to verify that the calculation techniques used for this work are accurate. Tensile modulus values of geotextiles at 5% strain and at failure based on information in Geotechnical Fabrics Report (1996). (1975) method is more conservative at these loading conditions.) MD 700/4000 542/3098 306/1749 7910/45206 876/5006 1402/8012 233/1333 500/2858 290/1657 10000/57150 267/1524 538. Stress distribution through the aggregate layer Figure 14 shows the soil strength vs.5/3078 XD 875/5000 783/4475 383/2186 667/3810 876/5006 1402/8012 300/1715 583/3334 310/1772 4167/23814 260/1486 487. 427/12 W/S-PP. aggregate thickness curves for both design techniques without geotextiles for dual wheels on a single axle with wheel loads of 60 and 115 kN (13.850 lb) and tire pressures of 414 kPa (60 psi). Details are given in Appendix B. Design curves provided in Barenberg et al. 4 × 4 TNS W300 USA Spantex 5710 Webtec. There is a difference between the curves generated for Figure 13 and those from Giroud and Noiray (1981) for the 450 kN/m geotextile being used for the 480 kPa tire pressure. These represent 10-ton and 20-ton trucks (e. mass/area (g/m2)/ (oz/yd2) W-PP.500 and 25. The reason for this discrepancy is unknown. Figures 12 are 13 are static load design curves from Giroud and Noiray (1981) and from calculations performed for this work. 200ST Synthetic Industries Gtx. Newmark 1942) and the trapezoidal stress distribution beneath a rectangular load used by Giroud and Noiray (1981).5/2786 Product Amoco 2044 Carthage FX-400MF Contech C-300 Huesker Comtrac 800 Linq GTF 550T Linq GTF 1000T Synthetic Industries Gtx. 150/4 W/C-PP. XD = cross-machine direction. TTHPG-50 Webtec. This difference is estimated to be about 10% at the very lowest values of aggregate thickness. 1430/42 W-PET. PET = polyester. Table 4 shows the maximum vertical stress at various depths below the load for a wheel load of 115 kN and contact pressure of 414 kPa using the Boussinesq stress distribution beneath a circularly loaded area (i. the geotextiles when they are being used and to relate them to the values measured in tensile tests. The Barenberg et al. (1975) and points calculated for this work to verify the calculations. (1975) used the Boussinesq stress distribution because experimental and field work of others show that stress distribution 13 .e. and this stems from the load distribution assumptions pertaining to the spreading of the load beneath the wheels.0 (Mathsoft 1995) to generate design curves. Table 1). Therefore. na W/S-PP. TTHPG-57 Notes: na = not available. 440/13 W-PP. na W-PP. Validation of calculation techniques Design equations were programmed using Mathcad 6. Similarly. the currently used Army design technique is compared with the design technique of Giroud and Noiray (1981) in this section.g.. 2566/76 W-PP. Construction.. 200/6 W-PET. MD = machine direction.Table 3. 203/6 K-PET.) MD 420/2400 386/2206 174/994 7200/41150 404/2309 1050/6000 174/994 384/2195 100/570 8000/45720 200/1143 700/4000 XD 760/4340 456/2606 210/1200 800/4572 404/2309 1050/6000 192/1097 454/2595 280/1600 4000/22860 220/1257 700/4000 K at failure (kN/m)/(lb/in. PP = polypropylene. Barenberg et al. W = woven. Figure 11 shows the static load design curves from Barenberg et al. Design curves for the loading imposed by typical military vehicles using each design method are also presented to demonstrate potential aggregate savings by use of the Giroud and Noiray (1981) method. COMPARISON OF GIROUD AND NOIRAY METHOD WITH ARMY METHOD The tensile reinforcement advantages offered by high-strength geotextiles may offset the increased cost. na W-PP K at 5% strain (kN/m)/(lb/in. K = knitted. na W-PET.

2 0 0 20 2 4 1 40 6 8 2 60 0 80 10 12 3 100 14 16 120 18 4 140 kPa 20 psi Undrained Shear Strength 0 CBR 0 Figure 12.in.6 } This Study Wheel Load = 89.8 0.2 kN (5. 14 .75 4 1. (1975) 40 Aggregate Thickness 1.000 lb) Tire Pressure = 480 kPa (70 psi) On-Highway Vehicle 30 20 0.0 kN (20. 50 m 1.000 lb) Tire Pressure = 345 kPa (50 psi) With Geotextile 0 Undrained Shear Strength 0 CBR 0 Figure 11.000 lb) Tire Pressure = 520 kPa (75 psi) Without Geotextile 30 20 0.00 30 5 1.000 lb) Tire Pressure = 620 kPa (90 psi) Off-Highway Vehicle Axle Load = 80 kN (18.2 0 0 10 1 0.2 Barenberg et al. (1975) Aggregate Thickness 40 } This Study Axle Load = 870 kN (196.0 0.0 0.4 1.25 40 6 1.4 10 0.6 Barenberg et al.2 1. Static loading design curves without geotextile from Giroud and Noiray (1981) and points generated for this report according to method documented in Appendix B.25 2 0.8 0. (1975) and design points generated for this report according to method documented in Appendix B.4 10 0. 50 m 1.50 7 50 kPa psi Wheel Load = 22. Static loading design curves from Barenberg et al. in.50 20 3 0.

0 35 30 Aggregate Thickness 0. 15 .900 lb) Barenberg wheel load = 60 kN (13.2 1.900 lb) Giroud & Noiray = 60 kN (13.2 0 0 20 2 4 1 40 6 8 2 60 10 80 12 3 100 14 16 4 120 18 140 kPa 20 psi 0 Undrained Shear Strength CBR 0 0 Figure 14.8 K = 10 kN/m.6 Barenberg wheel load = 115 kN (25. Tire Inflation Pressure = 621 kPa (90 psi) K = 450 kN/m. in.8 0.6 0. Static loading design curves with geotextiles from Giroud and Noiray (1981) and points generated for this report according to method documented in Appendix B.in. (1975) for 10.4 10 0. Static loading design curves adapted from Giroud and Noiray (1981) and Barenberg et al.and 20-ton trucks with tire pressures of 414 kPa (60 psi). m 1.4 Giroud and Noiray (1981) } This Study 0.500 lb) Aggregate Thickness 40 30 20 0.500 lb) Giroud & Noiray = 115 kN (25.0 0.2 5 0 0 10 2 1 20 4 30 40 6 2 50 8 3 60 70 10 80 12 90 14 100 kPa psi 0 Undrained Shear Strength 0 CBR 0 Figure 13.4 1. 50 m 1. Tire Inflation Pressure = 480 kPa (70 psi) 25 20 15 10 0.

Thus. 15). Newmark 1942).5 99.2/12.4 Stress according to Boussinesq (Newmark) method (kPa/psi) 400.5-m.* Depth below applied stress.(20-in. US Army Cold Regions Research and Engineering Laboratory.0 342.25 to 0.2/13.3 m (4 to 12 in.6 0. 1997. G.73 0.8 67.0/30.25-m.1/58. Fannin and Sigurdsson 1996.0 Stress according to trapezoidal stress distribution (kPa/psi) 275. Perloff 1975).0/10.58 0.2 0.7 151. 16 . mobility models also incorporate Boussinesq stress distributions.9 120.H. Research Civil Engineer.6-m. The ratio of the trapezoidal stress below a rectangle to the Boussinesq stress below a circular plate for these loading conditions ranges from 0.2 Ratio of the trapezoidal stress to the Boussinesq stress 0.. when a 45-kN (10kip) load was applied by an airplane tire at 690 kPa (100 psi).4 74.5 0.7 0. Blaisdell. and this should not be discounted as a potential factor in the observed unconservative design for static loading by the Giroud and Noiray method described above.2 73. The significant difference in estimation of stresses at the surface used by the two methods warrants further investigation. crushed limestone base rock layers ranging in thickness from 0.6 265.92 1.17 *The Boussinesq method used to generate results in this report did not add the pressure due to the weight of the overburden (= γ z) whereas the trapezoidal method used did.8 61.5) (e.61 for a 0.-) thick aggregate to 0. is recommended.* Although trapezoidal stress distribution below rectangularshaped loads is commonly used in shallow foundation design (e.1 0.4 0. which incorporates the Boussinesq stress distribution through the aggregate.9/9. McMahon and Yoder (1960) demonstrated that. The calculations were carried out in this manner to be consistent with how the original researchers presented them.4/39.9 1. from 30 to 40 kPa (CBR of about 1. use of the guidance in TM5-818-8.4/17.(24-in.. for compacted.3/9.9 116.1/21. Yoder and Witzak (1975) also refer to the use of a Boussinesq distribution of stresses below traffic loading for the purposes of pavement design.7/38.9/8. Indeed. If the weight of the overburden were added to the stresses estimated by the Boussinesq method.Table 4.0/21.69 0. until further investigation.60 0. Maximum vertical stress at various depths below applied wheel load of 115 kN and contact pressure of 414 kPa according to Newmark (1942) and trapezoidal stress distribution used by Giroud and Noiray (1981). through a granular layer to the subgrade follows the same pattern as that given by the Boussinesq theory.1/28.78 for the 0.9 91.(10-in.66 0.-) thick layer of sand was about twice that of the stress reaching the subgrade through a layer of crushed limestone (Herner 1955)..L.5 151.04 1.9 198.g.1/49. z (m) 0. N. Giroud and Noiray (1981) did not cite other work that uses trapezoidal stress distribution to estimate traffic loading stresses through aggregate.) and loaded with circular plates.4/7.3 0.7/14.6 49.57 0.-) thick aggregate (Giroud and Noiray 1981. the differences in stresses at depths of up to 1 m would be even greater than those listed in Table 5.6 59.6/8.82 0.5 210. the vertical stress reaching the subgrade through a 0. Fig.8 0. For example.2/16.g.1 to 0. Hanover. The aggregate quality significantly influences the stress distribution through it (Herner 1955).5 85.50 m thick and the subgrade strength ranges *Personal communication. There is limited evidence suggesting that the Giroud and Noiray (1981) method is unconservative for static loading conditions in both reinforced and unreinforced test sections when the aggregate layers are 0.0 57.8/10.

m 0. 17 .in. Barenberg et al. ) (2 (6 in 2 i n.1 15 0. (1975) design method.4 ) . Field performance vs.) r = 10 cm (4 in.) Base Course Thickness.) 1000 0 0 1 Figure 15.) r = 15 cm (6 in. h 0. 6 i n ( (2 m cm 5 c 15 r= r= 10 100 Field Trial: r = 5 cm (2 in.2 5 Su = 30 kPa (4. This would allow the confident use of design techniques for relatively low-quality aggregate that might be the only option for theater of operations military construction. and work that examines stress distributions through aggregates other than crushed rock should also proceed.) r = 15 cm (6 in. a Boussinesq stress distribution through it is a reasonable assumption for such an aggregate. based on the work of McMahon and Yoder (1960).1 . shapes other than a circularly loaded area should be considered.) cm c m m ( (6 in c 5 r = 1 5 r=5 cm r= 15 r= 0. al. (1975) based their theory on tests that utilized “crushed stone aggregate.) r = 10 cm (4 in. The possibility of using Boussinesq stress distribution through the aggregate layer could be added to the Giroud and Noiray (1981) design technique.5 Field Trial: r = 5 cm (2 in. Steward et al.8 psi) 0 0 0.3 10 0. theoretical prediction by Giroud and Noiray (1981) for unreinforced test sections (top) and reinforced test sections (bottom).5 15 0.) ) in.2 5 0.) r = 10 cm (4 in.) r = 5 cm (2 in.4 0.) the pressure measured in compacted clay-soil below the rock layers was reasonably approximated by a Boussinesq distribution beneath circular plates. (1977) did not describe the aggregate that was used in tests to validate the Barenberg et al.) in. In addition. .) Austin et.” and.4 psi) Su = 40 kPa (5. (From Fannin and Sigurdsson 1996.3 10 0. Unfortunately. et el.

the aggregate savings for the Giroud and Noiray (1981) method over the current Army method is about 0. For example. the aggregate savings for the Giroud and Noiray (1981) method over the current Army method is about 0. the geotextile is expected to undergo both biaxial tension and repeated loading.2 m (8 in.g. the portion on the outside of each set of wheels could easily slip into ruts formed by the vehicles. Design curves for the U. the load at the surface is not necessarily best modeled as a rigid circular area (as it is now).) was used for Figures 16 and 17 (bottom) because this value is easily obtained for commercially available products (Table 3). For the 20-ton truck at a soil strength of 40 kPa (5. Geotextile modulus values at 5% strain are readily available.S. field or other experimental work is needed to help establish the effective modulus values of the geotextiles when they are being used and to related them to the values measured in tensile tests. other wheel-load geometries should be considered. which is widely accepted and well-supported. However.Design curves for Army vehicles Because the potential for aggregate and cost savings is of interest to the U. less than 10 days of exposure) and aggregate is not available. Army. It is important to remember that the aggregate used with this method should have a minimum CBR of 80 (Giroud and Noiray 1981). Holtz et al. Finally. especially for expedient. Richmond et al. However. Army for reinforcement of thawing soils.. the tensile modulus values of some commercially available geotextiles far exceed those used in Figures 16 and 17 (bottom).4 psi). For the 10-ton truck. but possibly far lower for repeated loading than for monotonically loaded uniaxial tests— the tests that are now performed to determine geotextile modulus values. it is potentially of great interest to the U. For this concept to be implemented.. However.and 20-ton trucks are presented in Figures 16 and 17.. the length to width ratio for a HEMTT is estimated as L = 1. Future work should consider the use of available products with appropriately high modulus values. and the Giroud and Noiray (1981) method shows promise for large savings over the current Army design method. For use of a Boussinesq stress distribution. In reinforcing lowbearing-capacity soil. temporary operations where ultraviolet degradation due to exposure to sunlight is not a consideration (e. accounting for the tensile support provided by the geotextile provides considerable advantages of aggregate savings.8 psi). it should be further investigated because it promises large aggregate savings compared with the current design method.) with geotextile.6B (e. 1993). respectively. A geotextile tensile modulus of 200 kN/m (1143 lb/in.S. this appears to be a reasonable strain estimate for static loading of geotextiles performing reinforcement over low-bearing-capacity soils. design curves for Army vehicles were developed according to both methods for comparison of aggregate thickness required. due solely to its ability to account for tensile properties of the geotextile reinforcement at large rut depths.) with geotextile. Considerable aggregate savings can be realized if the Giroud and Noiray (1981) method is used. Based on limited field experiments. Depending on the outcome of an investigation of the stress distribution through the aggregate layer to the subgrade.g. modulus values are higher in biaxial tension. it may be worthwhile to develop a hybrid design method that uses a Boussinesq stress distribution through the subgrade with a membrane support mechanism as presented by Giroud and Noiray (1981).) An important factor in the adoption of the Giroud and Noiray (1981) design method for use is knowledge of the appropriate geotextile modulus values. Thus. the geotextiles would likely have to be anchored in some way in order for the tensile properties to fully develop and provide the necessary reinforcement. a situation that can be tolerated by military vehicles on thawing soils.2 m (8 in. The Giroud and Noiray (1981) design method indicates that the geotextile may be able to provide reinforcement with no aggregate on the surface. Therefore. RECOMMENDATIONS FOR FUTURE WORK Using the Giroud and Noiray (1981) method as it is presented herein may lead to unconservative design and construction results because the stress distribution through the aggregate layer to the subgrade is less than the Boussinesq method. Thus. (Even though the geotextile is in a state of tension between the wheels.g. this is not a currently recommended practice because of the increased risk of damage to the geotextile due to trafficking and because of deterioration when exposed to sunlight (e. Regardless of whether the Giroud and Noiray 18 . Army’s 10. 1990). with a soil strength of 30 kPa (4.S. As discussed earlier. This could result in substantial aggregate savings.

2 5 0 0 0 Undrained Shear Strength 0 10 2 20 4 30 40 6 50 8 60 70 10 80 12 90 100 14 CBR 0 1 2 3 4 Figure 16.in.8 Aggregate Thickness 25 20 15 10 0.4 With Geotextile 0. 19 .4 With 200 kN/m Geotextile 0. Use of upper figure is recommended until further research is conducted. Design curves for static loading (up to 100 passes) for 10-ton dump truck.6 Without Geotextile 0.0 0. m 1.2 5 0 0 10 0 Undrained Shear Strength 0 CBR 0 20 2 4 30 40 6 50 8 60 70 10 80 12 90 100 kPa 14 psi 1 2 3 4 in.6 Aggregate Thickness 25 20 15 10 0. 35 30 m 1. according to Barenberg (1975) method (top) and Giroud and Noiray (1981) method (bottom).0 35 30 0.8 Without Geotextile 0.

0 0.2 5 0 0 0 Undrained Shear Strength CBR 0 0 With 200 kN/m Geotextile 10 2 20 4 30 40 6 50 8 60 70 10 80 12 90 100 kPa 14 psi 1 2 3 4 Figure 17. according to Barenberg (1975) method (top) and Giroud and Noiray (1981) method (bottom).6 Without Geotextile 0 Undrained Shear Strength 0 CBR 0 1 2 3 4 in. 20 . Design curves for static loading (up to 100 passes) for 20-ton dump truck. Use of upper figure is recommended until further research is conducted. m 1.4 0. 35 30 Aggregate Thickness m 1.2 5 0 0 10 2 20 4 30 40 6 50 8 60 70 10 80 12 90 100 kPa 14 psi 0.in.6 25 20 15 10 0.4 With Geotextile 0.0 35 30 Aggregate Thickness 25 20 15 10 0.8 Without Geotextile 0.8 0.

122(7): 544– 553. and heliports in the theater of operations—Road design.P. St. These methods both require the use of high quality aggregate. December. p. Herner. 319–325. University of Illinois Report No. Geosynthetics. a method that accounts for repeated traffic loading is desirable. Army until further study is completed.F. Transportation Research Record 671. D. Headquarters. J. Task Force 25 (1990) Guide specifications and test procedures for geotextiles. 65–80. not just crushed rock. of the American Society of Civil Engineers Journal of the Geotechnical Engineering Division. Sigurdsson (1996) Field observations on stabilization of unpaved roads with geosynthetics. since the Giroud and Noiray (1981) design method may be unconservative.. even though soils are usually only temporarily in a weakened state when they thaw. considerable aggregate savings for the U. Kinney. these methods are likely to be conservative. T.. SUMMARY AND CONCLUSIONS Table 1 may be used with the design curves presented in TM5-818-2 for convenience in using the current Army geotextile reinforcement design method.or 20-ton trucks are expected to exert the maximum wheel loads on thawing or other low-bearing-capacity subgrade soils. September. If the thawed layer is less than 0. p. and J. American Society of Civil Engineers Journal of Geotechnical Engineering.M. and this should also be included in future development efforts. Thus. D. Giroud. and O.C.S. Noiray (1981) Geotexile-reinforced unpaved road design. (1983) Jane’s Military Vehicles and Ground Support Equipment.. D.J. (1992) Biaxial tensile state of stress in geosynthetics. Barenberg (1978) Design and behavior of soil–fabric–aggregate systems. Journal of the Geotechnical Division. D. Coleman (1993) A field evaluation of geosynthetic-reinforced haul roads over soft foundation soils. Fannin. the use of a variety of aggregates.N. airfields. FM5-430-001/AFJPAM 32-8013. E. American Society of Civil Engineers. respectively. More rounded material such as sand and gravel has been shown to concentrate stresses over a significantly smaller area on the subgrade than crushed rock. Proceedings. (1982) Discussion of geotextile-reinforced upaved road design. Alexandria. However. UILUENG-75-2020. C. LITERATURE CITED AASHTO-AGC-ARTBA Joint Committee: Subcommittee on New Highway Materials. A hybrid method.) thick. Austin. and repeated traffic loading. Vol. Washington. Soil and Rock (II): D4943–latest. if 10.J. 1998 Annual book of ASTM Standards. prepared for Celanese Fibers Marketing Company.P. Industrial Fabrics Association International. this would be an important addition to the current design method. Minn.J.C. American Association of State Highway and Transportation Officials. Figure 16 or 17 (top) may be used. Geosynthetics 93. Washington. 108(GT12): 1657–1658.. Giroud.09. However. Paul. Vol. 11. should also be completed so the design method can be adjusted accordingly. Geotextiles and Geomembranes. If further research proves that the Giroud and Noiray (1981) design method is adequate. and D.A. Foss. 64–75.C. . Fourth edition. J. Army would be realized by using it. J. Vol. it should not be used by the U. they will sometimes have to carry more than 100 vehicles during thawing. and L. estimates of stress through a variety of aggregates. Dowland (1975) Evaluation of soil–aggregate systems with MIRAFI fabric. However. 4. it should also include determination of representative modulus values. Hales.(1981) design method is selected for use by the U. p. In Proceedings. Due to the likelihood that theater of operations construction will be completed with limited sources of high-quality aggregate.C. Army.4 m (16 in. Bender. Department of the Air Force. Va. combining a Boussinesq stress distribution through the aggregate layer with a membrane support mechanism as presented by Giroud and Noiray (1981) might be an optimum design technique. Barenberg. the shape of the wheel load. July. a geotextile separator will probably still provide benefit to lengthen times between maintenance of the gravel surface. ASTM D 4595-86 (1998) Standard test method for tensile properties of geotextiles by the wide-width strip method. R. 107(GT9): 1233–1254. Finally. and E. When this approach is further developed.S. Department of the Army. (1955) Effect of base course quality on load transmission through flexible pavement. 1 (1994) Planning and design of roads.S. R. 21 ..

H. Washington.C. R. Highway Research Board. Yoder (1960) Design of a pressure-sensitive cell and model studies of pressure in a flexible pavement subgrade. (1975) Pressure distribution and settlement. Cambridge. Mohney (1977) Guidelines for use of fabrics in construction and maintenance of low-volume roads. Witczak (1975) Principles of Pavement Design.J.. Geotextiles and Geomembranes. In Proceedings.W. N.0. p. Englewood Cliffs.. p..Y. Division of Engineering. U. Fang. W. Mathsoft (1995) Mathcad PLUS 6.D. T. Williamson and J. FHWA HI-95038. and M. 224–233. Holtz. Soderman. and E.. 2(2): 495–504. Christopher and R. Vol. Vol. D.R. and W.E.C. 5. K. 40. Foundation Engineering Handbook. and J. 650– 682. Steward. N.D.J. 148–196.F. Headquarters.S.: Prentice-Hall.P.F.. Portland. Chapter 4. New York: Van Nostrand Reinhold. Washington. US Forest Service. Second edition. D. eds. McMahon. Department of Transportation. Yoder. p. R. Departments of the Army and the Air Force. H.. E.. Geosynthetics International. p. University of Illinois Bulletin. MathSoft Inc. Oregon.R. J..J. 191–199. Inc. 1955. Highway Research Board. Giroud (1992) Relationship between uniaxial and biaxial stresses and strains in geosynthetics. Berg (1995) Geosynthetic design and construction guidelines.D. B. New York: John Wiley and Sons. Kovacs (1981) An Introduction to Geotechnical Engineering. Newmark. Holtz. Winterkorn and H. Mass. Holtz. 1960. R.M.D. Perloff. R. 22 . TM5-818-8/AFJMAN 32-1030 (1995) Engineering use of geotextiles. Sivakugan (1987) Design charts for roads with geotextiles. 12. No. Participant notebook.In Proceedings. (1942) Influence charts for com- putation of stresses in elastic foundations. Federal Highway Administration. and N.L.

Task Force 25 (1990) and is more recent guidance than that provided in TM5-818-8. AOS <0. cValues of geotextile elongation do not imply the allowable consolidation properties of the subgrade soil. so that geotextiles meeting survivability requirements may be specified for acquisition without referring to TM5-818-8 (1995). Geotextile physical property requirements for survivabilitya <50% elongation/>50% elongationb. >50% soil finer than 0. Army TM5-818-8. Site soil CBR Equipment ground contact pressure. It can be used in place of Tables 2-2.c 150/6 300/12 450/18 aMaximum bFor <1 >350(50) <350(50) 1 to 2 >350(50) <350(50) >3 >350(50) <350(50) NR NR NR H NR NR H M H H M M H H M M M M M M M M M M aggregate size not to exceed one half of compacted cover thickness. Table A-2. NR = not recommended.APPENDIX A: GEOTEXTILE SURVIVABILITY REQUIREMENTS This appendix is provided as a convenience to readers. M = medium. AOS > 0.075 mm. and 2-4 in U. Construction survivability ratings. cMinimum cover thickness is limited to existing road bases and is not intended for use in new construction. low-volume. Table A-1.075 mm. as determined by ASTM D 4632. compacted (mm/in. <50% soil finer than 0.S.)a 100/4b. 70% of strength retained for all cases Geotextile acceptance aValues Required survivability level H M Puncture resistance (kN/lb) ASTM D 4833 100/75 70/40 Trapezoidal tear strength (kN/lb) ASTM D 4533 100/75 70/40 Test method ASTM D 4751 ASTM D 4491 ASTM D 4355 ASTM D 4759 shown are minimum roll average values. kPa (psi) Cover thickness. The guidance provided here is taken from AASHTO-AGC-ARTBA Joint Committee: Subcommittee on New Highway Materials. H = high. 2-3.6 mm 2. unpaved roads (average daily traffic less than 200 vehicles). These must be determined by a separate investigation. bElongation 23 .c Grab strength (kN/lb) ASTM D 4632 270/180 180/115 Additional requirements Apparent opening size 1.3 mm Permeability Kgeotextile > Ksoil Kgeotextile = permittivity × nominal geotextile thickness Ultraviolet degradation At 150 hr exposure.

24 .) 229/9 203/8 76/3 — NR = Not recommended. Soil strength (CBR) <1 1–2 2–3 >3 Unsewn overlap (mm/in.) NR 965/38 762/30 610/24 Sewn overlap (mm/in. Recommended overlaps.Table A-3.

  For a range of soil cohesion. (1975) necessitated the solving of equations for the aggregate thickness required. equations for cohesion were solved for a given range of aggregate thickness.g. given a known range of soil cohesion.0 software used to make these calculations follow. and r is the radius of the circle and z is the depth at which the stress determination is desired (z is located directly below the center of the circularly loaded area). The design method presented by Barenberg et al. z. the stress that can be tolerated at the subgrade is given by either 3.. (1975) assumed that the stress transmitted to the subgrade surface through the aggregate layer can be approximated by a Boussinesq stress distribution through an elastic.0 c with geotextile. isotropic solid bounded by a plane horizontal surface is (Newmark 1942) σ = 1 – cos3 α (B. isotropic half-space. and a known applied stress.0) to yield 1  3  (1 – σ ) z = r 2  1 – (1 – σ ) 3    . Thus. Mathcad 6. SOLUTION FOR DEPTH OF AGGREGATE FOR BARENBERG DESIGN METHOD Barenberg et al.0 (Mathsoft 1995) was used to solve these equations and generate design curves as well as to produce a symbolic equation for the aggregate thickness required for the Barenberg et al. c. AND GIROUD AND NOIRAY.3 c (without geotextile) or 6. Example work sheets from the Mathcad 6. to the stress on a uniformly loaded circular area in an elastic. (1975) approach.1 and is rearranged to yield 3  r a tan  = a cos(1 – σ ) . 25 . The expression for α is substituted into eq B.1) where α = a tan(r / z) . homogeneous.3c/contact pressure) is used to determine the thickness of the aggregate layer needed.APPENDIX B: METHODS USED TO RECALCULATE DESIGN CURVES FROM BARENBERG ET AL.2) This equation is then solved for z (using Mathcad 6. For Giroud and Noiray (1981). homogeneous.  z 1 (B. e. the ratio of soil strength to applied pressure. σ = (3. The ratio of vertical stress at depth.

43709 m 0.17497 m [ ] 2 3 c 1.35113 5.0c ConPress Stuff in English units: 1.0 WORKSHEET FOR BARENBERG DESIGN METHOD Solving of Boussinesq equation for z.80151 7.58732 m 0. Wload = 115 000 newton ConPress = 414 000 Pa r = 0.000 000 104 kg m–1 sec–2 This part of the sheet creates a database and makes a .prn file out of it for importing to a spreadsheet later.000 000 104 kg m–1 sec–2 3. With geotextile.r) 26 .33620 m 0.EXAMPLE MATHCAD 6.70226 1 c psi r = 0.000 000 104 kg m–1 sec–2 2.25473 m 0.89647 m 0. applied contact pressure is 414 kPa.000 000 104 kg m–1 sec–2 4.97557 ft z(c.45038 2.000 000 104 kg m–1 sec–2 6. WRITEPRN (Baretwent) = z(c.25189 8.000 000 104 kg m–1 sec–2 5. 20 000 Pa … 60 000 Pa σ( c ) = 6. the depth of aggregate for wheel loads of 60 and 115 kN.90075 4.29735 m r= Wload ConPress ⋅ π c = 10 000 Pa.r ) = r [1– σ](c) 3 1– 1– σ(c) z(c.r) 0.

EXAMPLE MATHCAD 6. 36. 30 and 31 as well as 5 and 7 (onhighway trucks). 5: h = 0. meaning that the parabola between wheels is wider than the sum of the widths of the parabolas under the wheels). It uses eq 43.5(e – B – 2 · h · tan α) Settlement of geotextile from original position. 35.3 m Width of wheel load (on road). 36:     2 2   2 ⋅ s(h)   2 ⋅ s(h)   a(h)  2 ⋅ s(h)   b(h) = a(h)1 + 0.6 Pc = 414 000 P e = 2. K = 200 000 N m–1 tan α = 0.527 m Width of parabola between wheels. 30: a(h) = 0. The original reference is Giroud and Noiray (1981).0 m B= P Pc B = 0. 31: aprime(h) = 0.5 1 +  ⋅ ln + 1+   +   – 2  2 ⋅ s(h)   a(h)   a(h)                 Equation for half length of parabola between the wheels.0 WORKSHEET FOR GIROUD AND NOIRAY DESIGN METHOD This is a file to calculate the aggregate depth needed for different K values of the geotextile. 35:  b(h) + bprime(h)  e(h) =   –1  a(h) + aprime(h)  27 .1 … 1.0 m P = 230 000 newton r = 0. and 37 (a' > a. 37:        2 ⋅ r – s(h) bprime(h) = aprime(h) ⋅ 1+ 0. 0.5 1+     aprime(h)         aprime(h) r – s(h)  + ⋅ ln 2 ⋅ +  aprime(h)   2 ⋅ r – s(h)     2 [ ]   [ ]  2 ⋅ r – s(h)  + 1+   aprime(h)  [ ]       – 2             2   The elongation of the geotextile. 33.5 (B + 2 · h · tan α) L= B 2 L = 0. 33: s(h) = r ⋅ aprime(h) a(h) + aprime(h) Equation for half length of parabola under the wheel.745 m Length of wheel load(on road). 7: Width of parabola under wheel.

181 104 kg m–1 sec–2 3.2 m 0.8 m 0.116 103 kg m–1 sec–2 6.36 103 kg m–1 sec–2 8.058 103 kg m–1 sec–2 7.111 104 kg m–1 sec–2 9. P. soil strength. 43:       K ⋅ ε(h) 1  P  Cu(K.1 m 0.P.3 m 0.346 104 kg m–1 sec–2 1. h) 5.4 m 0.9 m 1m 28 .h) = ⋅ –  π + 2  2 ⋅ (B + 2 ⋅ h ⋅ tanα ) ⋅ (L + 2 ⋅ h ⋅ tan α ) 2  a(h)   a(h) ⋅ 1+      2 ⋅ s(h)     Cu (K.5 m 0. P.7 m 0.663 104 kg m–1 sec–2 1. h) h 0m 0.477 103 kg m–1 sec–2 WRITEPRN(Gir20t) = Cu (K.659 104 kg m–1 sec–2 2.099 104 kg m–1 sec–2 1.And the equation of thickness of aggregate vs.6 m 0.722 104 kg m–1 sec–2 2.

Henry 7. 14. To stabilize the soil. PERFORMING ORGANIZATION REPORT NUMBER U.REPORT DOCUMENTATION PAGE 1. DISTRIBUTION/AVAILABILITY STATEMENT 12b.. and to the Office of Management and Budget. the current Army design technique should be used until more research is conducted. Design involves selection of aggregate thickness and geotextile. In the meantime. Future work should consider adopting a hybrid design method that provides realistic estimates of stresses at the subgrade and accounts for the tensile properties of geotextiles. searching existing data sources. SUBJECT TERMS Geosynthetics Geotextiles Low-bearing-capacity soils 18. and completing and reviewing the collection of information. distribution is unlimited. Z39-18 298-102 . PRICE CODE 17. including suggestion for reducing this burden. Paperwork Reduction Project (0704-0188). Directorate for Information Operations and Reports. AGENCY USE ONLY (Leave blank) 4. Suite 1204. a geotextile is placed on it. DC 20503. 1215 Jefferson Davis Highway. including the time for reviewing instructions. Virginia 22161 13. ABSTRACT (Maximum 200 words) Thawing fine-grained soils are often saturated and have extremely low bearing capacity. DC 20314-1000 10. up to 100 vehicle passes) are presented and compared in this report. and the Army uses one of them. SECURITY CLASSIFICATION OF THIS PAGE Military vehicles Reinforcement Thawing soils 19. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. FUNDING NUMBERS June 1999 Geotextile Reinforcement of Low-Bearing-Capacity Soils: Comparison of Two Design Methods Applicable to Thawing Soils 6. 3. REPORT TYPE AND DATES COVERED 5. to Washington Headquarters Services. LIMITATION OF ABSTRACT UNCLASSIFIED NSN 7540-01-280-5500 UNCLASSIFIED UNCLASSIFIED UL Standard Form 298 (Rev. REPORT DATE Form Approved OMB No.e. TITLE AND SUBTITLE 2. this alternative method appears to be unconservative with respect to stresses estimated at the subgrade surface. aggregates other than the high-quality crushed rock that is inherently assumed by each design method should be accounted for in new design development.S. NUMBER OF PAGES 36 16. straightforward design curves for Army 10. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) Special Report 99-7 Office of the Chief of Engineers Washington. saturated soils and therefore are good candidates for use in stabilization of thawing soils. 2-89) Prescribed by ANSI Std. gathering and maintaining the data needed. The theory and use of the two design methods for static loading (i. VA 22202-4302. Army Cold Regions Research and Engineering Laboratory 72 Lyme Road Hanover. Springfield. SPONSORING/MONITORING AGENCY REPORT NUMBER 11. There are two commonly used design techniques for geotextile reinforcement of low-volume roads. However.and 20-ton trucks as well as vehicle loading and tire pressure information for a number of other vehicles are included in this report to help make the current design method easy to use. then the geotextile is covered with aggregate. Arlington. SECURITY CLASSIFICATION OF REPORT 20. SUPPLEMENTARY NOTES 12a. Available from NTIS. SECURITY CLASSIFICATION OF ABSTRACT 15. New Hampshire 03755 9. Send comments regarding this burden estimate or any other aspect of this collection of information. DISTRIBUTION CODE Approved for public release. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response. Geosynthetics are used to reinforce unsurfaced roads on weak. In addition. AUTHORS WU: TO-007 4A 161102AT4Z Karen S. The design method not used by the Army offers the potential to reduce aggregate thickness over the geotextile because it accounts for the fact that the geotextile helps support the traffic load (when in tension) and confines the soil between the wheels and the subgrade. Thus. Washington.