Anda di halaman 1dari 33

CHAPTER NINE

Ground properties: rock strength


and compressibility

Introduction
This chapter considers ground properties because they are the key to
understanding ground behaviour and performance. It is important to the
civil engineer to know how the ground will react to construction being
dug, built on, and used as construction materials and how these
properties may change with time.
It is important to distinguish between soil and rock because their
properties are very different and so their engineering behaviour is too.
This has given rise to different disciplines for analysing stability and
settlement of ground. Civil engineers, on the one hand, learn soil
mechanics where ground is idealised as a continuum within which failure
mechanisms such as slip surfaces can spontaneously develop (e.g. in a
cut slope or an embankment foundation slipping on a circular arc or
some other mechanism), as well as pre-exist. Pore water pressure is dealt
with implicitly in terms of effective stress, the difference between total
stress and pore water pressure, which is a measure of the loading transmitted by the structural skeleton of soil grains. Engineering geologists
and mining engineers, on the other hand, learn rock mechanics where
ground is regarded as an assembly of rigid blocks that slide (or not) along
existing frictional joints or surfaces that may or may not be orientated to predispose a collapse of the rock mass, for example by sliding out of the side of
an excavation like a tunnel, cutting or quarry. Water pressures from water
filling the joints are dealt with as hydrostatic force vectors.
Soil is made up of particles of weathered rock that are microscopic in the
case of clays or visible to the naked eye in the case of sands. Not surprisingly,
soil behaviour is governed by this particulate character. Unlike rocks where
particles are bound by being crystalline or cemented, in soils the particles
are free to move by rolling, crushing and changing their packed structure
from loose to dense (consolidation or compaction) or dense to loose (swelling
or dilantancy). The interstitial voids or pores between particles interconnect
and provide flow paths for groundwater to flow through (very slowly in the
case of clays, more rapidly in the case of sands). Engineering behaviour and

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


the key properties of soil strength and stiffness are therefore governed by
the sizes and distribution of sizes of particles and their density of packing.
Conversely, rock behaviour is governed by discontinuities which
separate the rock mass into an assembly of rigid blocks. The overall engineering performance of the rock mass will be dictated by the number and
distribution of the discontinuities and crucially their compressibility
which is very much lower than the intact rock, in contrast to soil where
compressibility of fissures is not so markedly lower than the intact soil.
It is therefore the structure of geomaterials (soil and rock) that controls
the engineering behaviour of the ground. Take, for example, a landslip.
In the case of soil, if the bedding of the soil follows the downslope of a
potential slip, long translational slips can occur. If the soil structure is
random or the material homogeneous, failure can occur on a circular arc
with a rotational slip. In the case of rock, however, slips or topples can
occur only if the jointed block structure allows a feasible mechanism.

Rock properties and behaviour


Overview
Cohesive or fine-grained soils such as clays are highly time-dependent in
their behaviour. For example, the recovery of pore pressures in some
Victorian railway cuttings in the London Clay took over half a century,
leading ultimately to collapse. Rocks, on the other hand, are not timedependent in their behaviour over the life of a civil engineering structure.
Rocks may change their properties due to weathering but the time-scale
involved will normally greatly exceed the design life of a foundation
(typically 5090 years).
Key terminology
The engineering behaviour of rocks is almost always overwhelmingly conditioned by the rock discontinuities such as joints, faults and fractures.
These discontinuities are planes of weakness across which there is little
or no tensile strength. Mechanisms of excavation collapses, land slips and
bearing capacity failures will be feasible or not and will activate depending
on the extent, pattern and types of discontinuity present in the rock mass.
The mass compressibility of the rock is a combination of the compressibility
of the intact rock and the compressibility of the joints. Clearly, the greater
the extent of the joints and the lower their compressibility, the more the
joint properties will dominate overall mass behaviour.
Stiness of rock masses
For civil engineers, it is more usual to express the deformability of rock
(and soil too) in terms of the material stiffness, Youngs Modulus, E, the
ratio of stress increment to consequent strain increment in a uniaxial or
220

CHAPTER 9 GROUND PROPERTIES

1
Ej/Ei = 0.5

Em/Ei

0.8
Ej/Ei = 0.05

0.6

Ej/Ei = 0.0001

0.4

Ej/Ei = 0.005
Aperture = 1.0 mm

0.2
0

Ej/Ei = 0.0005
0

12
16
20
No. of joints per metre

24

28

(a)

Em /Ei

0.8

10 mm

0.6
Aperture
0.5 mm

0.4
0.2
0

Ej/Ei = 0.005

5 mm
20 mm
0
4

2 mm

12
16
20
No. of joints per metre

1 mm

24

28

(b)

Fig. 9.1 (a) Variation in the ratio of mass stiffness to intact stiffness with
fracture frequency for fractures with different stiffnesses; (b) variation in
the ratio of mass stiffness to intact stiffness with fracture frequency for
fractures with different apertures (after Matthews, 1993; figures prepared
by Marcus Matthews)
triaxial test. Stiffness has units of stress, MPa. Compressibilty is the
inverse of stiffness and has units of MPa1 .
The way discontinuities can dominate stiffness of rock masses is
illustrated in Fig. 9.1. Here, Matthews (1993) used the formulation of
Hobbs (1975) to carry out a parametric study of the effect on mass stiffness
of fracture (or joint or discontinuity) frequency and fracture thickness or
aperture. In Fig. 9.1, the subscripts correlation is m for mass (i.e. including joints or fractures), j for joint (i.e. joint properties only), and i for
intact (i.e. intact or parent rock only).
The case with the joints open to 1.0 mm, aligned perpendicular to the
direction of the applied load is shown in Fig. 9.1(a). This represents a
geometry often found in chalk. It will be seen from Fig. 9.1(a) that as the
modulus of the joint approaches that of the intact material, i.e.
Ej =Ei 0:5, the number of joints has little effect on the ratio of mass

221

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


modulus to intact modulus. The implication is that for the softer type of
materials such as soils, the joint system does not cause an appreciable
difference between the mass modulus and the intact modulus. The most
significant feature of this model, however, is the extremely rapid drop in
Em =Ei with the introduction of only a few (1 to 8) fractures per metre
when the ratio of Ej =Ei is less than 0.005 (or 0.5%). When the fracture
frequency exceeds about 10 per metre, the mass modulus becomes
relatively insensitive to increasing number of joints.
The relationships shown in Fig. 9.1(a) assume a constant joint aperture
of 1.0 mm. If the ratio Ej =Ei is kept constant, and the aperture varied, the
set of curves shown in Fig. 9.1(b) results. Not surprisingly, the greater
the aperture, the greater the mass compressibility. Again, a small
change in the variable (of only 1 or 2 mm of aperture) can bring about a
dramatic change in mass stiffness.
So of what use to the civil engineer is the intact stiffness, Ei ? This can be
measured in the rock mechanics laboratory by a uniaxial or triaxial test.
This can serve as an indicator of the quality and strength of the parent
rock only but not the rock mass including the often dominating joints. In
order to determine the mass stiffness, however, requires a full-scale or
near full-scale (i.e. representative size with respect to joint spacing and
aperture) test like a plate loading test (Matthews, 1993). Alternatively,
cross-hole and down-hole transmission of shear waves can measure
maximum shear modulus averaged over the distance traversed, i.e.
including joints (Simons et al., 2002).
In most civil engineering practice, however, stiffness and engineering
quality is determined by using a classification system for jointed rock
masses. These are discussed below.

Interpretation of rock properties (Mayne et al., 2002)


Source
This section is extracted from Chapter 10 Interpretation of rock properties of the Manual on Subsurface Investigations published by the
National Highway Institute of the Federal Highway Administration,
written by Paul W. Mayne, Barry R. Christopher and Jason DeJong.
This is published online. The full reference is as follows:
Mayne, P. W., Christopher, B. R., Berg, R. R., and DeJong, J. (2002). Subsurface Investigations Geotechnical Site Characterization. Publication
No. FHWA NHI-01-031, National Highway Institute, Federal Highway
Administration, Washington, DC.

Introduction
The engineering behaviour of most rock masses under loading is determined primarily by the discontinuities, fractures, joints, fissures, cracks
222

CHAPTER 9 GROUND PROPERTIES


and planes of weakness. The intact blocks of rock between the discontinuities are usually sufficiently strong, except in the case of weak and porous
rocks and those that weather rapidly. Thus, two classification systems are
needed to adequately characterise these geomaterials: one for the intact
solid rock and another for the rock mass. The network of fractures
divides the rock mass into discrete and prismatic blocks that affect its
response and performance. With the exception of the durability testing,
the results of laboratory testing are of limited direct applicability to
design of structures founded in or on rock masses.
Of the three primary rock types (igneous, metamorphic and sedimentary),
sedimentary rocks comprise 75% of the rocks exposed at the ground
surface. Among the sedimentary rocks, the rocks of the shale family (clay
shale, siltstone, mudstone and claystone) predominate, representing over
50% of the exposed sedimentary rocks worldwide (Foster, 1975).
An initial step during site reconnaissance and exploration is to
categorise the basic type of rock (Table 9.1). Detailed geological classifications of rock types and petrographic examinations in the laboratory will be
required for major projects involving construction on rocks. Field mapping
by engineering geologists is necessary for description of the jointing
patterns, major discontinuity sets, shear zones and faults, particularly in
areas involving rock slopes, cliffs, tunnels and bridge abutments. A
detailed discussion of these aspects may be found elsewhere (Pough,
1988; Goodman, 1989). Major slip planes and joints should be detailed
on maps with appropriate values of dip angle and dip direction (or alternatively, strike). Large groups of discontinuities are best represented by
statistical summaries on stereonets and polar diagrams. Important shear
zones and faults can also be depicted on these plots.
Alternate classification systems are proposed based on behavioural
aspects (Goodman, 1989) or composition and texture (Wyllie, 1999).
Details on the specific rock minerals and their relative abundance are
Table 9.1 Primary rock types classified by geologic origin (from Mayne et al., 2002;
permission of Federal Highway Administration)
Grains
aspects

Sedimentary types

Metamorphic types

Igneous types

Clastic

Carbonate

Foliated

Massive

Intrusive

Extrusive

Coarse

Conglomerate
breccia

Limestone
Gneiss
conglomerate

Marble

Pegmatiite
granite

Volcanic
breccia

Medium

Sandstone
siltstone

Limestone
chalk

Schist
phyllite

Quartzite

Diorite
diabase

Tuff

Fine

Shale
mudstone

Calcareous
mudstone

Slate

Amphibolite Rhyolite

Basalt
obsidian

223

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


Table 9.2 Geologic time-scale (from Mayne et al., 2002; permission of
Federal Highway Administration)
Era

Period

Epoch

Cenozoic

Quaternary

Holocene recent
Pleistocene
Pliocene
Miocene
Oligocene
Eocene
Paleocene

Tertiary

Mesozoic

Cretaceous
Jurassic
Triassic
Permian

Paleozoic

Carboniferous
Devonian
Silurian
Ordovician
Cambrian

Time boundaries
(years ago)
10 000
2 million
5 million
26 million
38 million
54 million
65 million
130 million
185 million
230 million
265 million

Pennsylvanian
Mississippian

310 million
355 million
413 million
425 million
475 million
570 million

Precambrian
(oldest rocks)

3.9 billion

Earth beginning

4.7 billion

important in the petrographic determination of the rock types, yet beyond


the scope of discussion here. In the logging of field mapping and rock
coring operations, the specific formation name and age of the rock is
often noted, being helpful in sorting stratigraphic layering and the determination of the subsurface profile. Table 9.2 gives the general geologic
time-scale and associated periods. Generally, older rocks have lower porosity and higher strength than younger rocks (Goodman, 1989).
Rock type can often infer possible problems that can be encountered in
construction. Notable problems occur in limestone (sinkholes, caves),
serpentine (slippage), bentonitic shales (swelling, slope stability) and
diabase (boulders). Deterioration of the shale family of rocks and
weakly cemented friable sandstones is the cause of many of the maintenance problems in the US national highway system, particularly with
respect to cuts, embankment construction and foundations. For example,
deterioration of cut slopes in shales will result in flatter slopes and/or
instability. Shale used in embankments when compacted will break
down and result in a material less pervious than anticipated for a rock
fill. Maintenance problems for slopes can be mitigated by making them
flatter, installation of horizontal drains, use of gunite and mesh, or in

224

CHAPTER 9 GROUND PROPERTIES


some cases, more elaborate structural supports are required (rock bolts,
retention walls, anchors, drilled shafts). When excavation for a structural
foundation is made, the bearing level must be protected against slaking
and/or expansion; this can be accomplished by spraying a protective
coating on the freshly exposed rock surface, such as gunite or shotcrete.
Additional details and considerations are given in Wyllie (1999).
The design of rock structures is still frequently done on the basis of an
empirical evaluation of rock mass properties guided by experience, consideration of rock mass structure, index properties and correlations, and
other parameters, such as joint spacing, roughness, degree of weathering,
dip and dip direction of slip planes, infilling, extent of discontinuities and
groundwater conditions (Fig. 9.2). Many of these facets can be grouped
together to give an overall rating of the predominant factors affecting

K: number of sets
B, Jq, Jv

E: orientation
a dip direction
y dip

I: spacing
S1 = Sapp sin q Sbedding
y dip

J1

Jv
S1

A: rock type

G: aperture (open)

Sbedding
M: seepage
J: persistence (l)
l
L: block size/shape

B: wall strength

Sapp

H: filling type
width

F: roughness

D: discontinuity type
bedding, fault etc.

Fig. 9.2 Factors and parameters affecting geologic mapping of rock masses features
(Wyllie, 1999 from Mayne et al., 2002; permission of Cengage)

225

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


the performance of the entire rock mass under loading. Thus, a rating of
the rock mass will be described using three common methods
(RMR rock mass rating; Q system and GSI geologic strength index).
As in the case of the evaluation of soil properties, a number of correlations have been developed for the interpretation of rock properties.
Notably, however, the rock property correlations reported in the technical
literature often have a limited database and should be used with caution.
An attempt should be made to develop correlations applicable to the
specific rock formations in a particular state, as this can be well worth
the expenditure of time and effort in terms of overall safety and economy.
This chapter presents general discussions on the properties of intact
rock and jointed rock masses, particularly using rock mass classification
schemes and their relevance to the design of rock structures. Civil
engineers are strongly encouraged to refer to the original references to
understand the basis of the correlations and the classification systems
presented in this chapter, and also for additional information.

Intact rock properties


This section presents information on the indices and properties of natural
intact rock. The values are obtained from tests conducted in the laboratory
on small specimens of rock and therefore must be adjusted to full-scale
conditions in order to represent the overall rock mass conditions.
Specic gravity
The specific gravity of solids (Gs ) of different rock types depends upon the
minerals present and their relative percentage of composition. The values
of Gs for selected minerals are presented in Fig. 9.3. Very common
minerals include quartz and feldspar, as well as calcite, chlorite, mica
and the clay mineral group (illite, kaolinite, smectite). The bulk value of
Galena
Pyrite
Barite
Olivine
Dolomite
Calcite
Chlorite
Feldspar
Quartz
Serpentine
Gypsum
Halite

Common minerals
Average Gs = 2.70

Reference value
(fresh water)

Specific gravity of solids, Gs

Fig. 9.3 Specific gravity of solids for selected rock minerals (from Mayne
et al., 2002; permission of Federal Highway Administration)

226

CHAPTER 9 GROUND PROPERTIES


Table 9.3 Representative range of dry unit weights (from
Mayne et al., 2002; permission of Federal Highway
Administration)
Rock type

Unit weight range: kN/m3

Shale
Sandstone
Limestone
Schist
Gneiss
Granite
Basalt

2025
1826
1927
2328
2329
2529
2030

1. Dry unit weights are for moderately weathered to unweathered rock.


Note: 9.81 kN/m3 62.4 pcf.
2. Wide range in unit weights for shale, sandstone and limestone represents
effect of variations in porosity, cementation, grain size, depth and age.
3. Specimens with unit weights falling outside the ranges contained herein
may be encountered.

these together gives a representative average value of Gs  2:7  0:1 for


many rock types.

Unit weight
The unit weight of rock is needed in calculating overburden stress profiles
in problems involving rock slopes and tunnel design support systems.
Also, because the specific gravity of the basic rock-forming minerals
exhibits a narrow range, the unit weight is an indicator of the degree of
induration of the rock unit and is thus an indirect indicator of rock
strength. Strength of the intact rock material tends to increase proportionally to the increase in unit weight. Representative dry unit weights for
different rock types are contained in Table 9.3.
The dry unit weight (dry ) is calculated from the bulk specific gravity of
solids and porosity (n) according to:
dry water Gs 1  n

7:1

where the unit weight of water is water 9:81 kN/m3 . The saturated unit
weight (sat ) of rocks can be expressed as:
sat water Gs 1  n n

7:2

The interrelationship between porosity and void ratio (e) is simply


n e=1 e. The decrease in saturated unit weight with increasing
porosity is presented in Fig. 9.4 for various rocks and a selected range of
specific gravity values.

Ultrasonic velocities
The compression and shear wave velocities of rock specimens can be
measured in the laboratory using ultrasonic techniques. These wave

227

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


28

sat = water[Gs(1 n) + n]
Saturated unit weight, T: kN/m3

26
Gs = 2.80, 2.65, 2.50
24
22
20
Dolostone
Greywacke
Mudstone
Sandstone

18
16

Granite
Limestone
Siltstone
Tuff

14
0

0.1

0.2

0.3

0.4

0.5

0.6

Porosity, n

Fig. 9.4 Saturated rock unit weight in terms of porosity and specific gravity
(from Mayne et al., 2002; permission of Federal Highway Administration)

5000
4500

Shear wave, Vs: m/s

4000
3500
3000
2500
2000
1500
1000
500
0
0

1000

2000

3000

Limestone
Tuff
Diorite
Basalt

4000
5000
6000
Compression wave, Vp: m/s
Chalk
Slate
Gabbro
Dolostone

Marble
Anhydrite
Granite
Mudstone

7000

8000

9000

10 000

Schist
Grandiorite
Dunite
Siltstone

Fig. 9.5 Representative S- and P-wave velocities for intact rock materials (from Mayne
et al., 2002; permission of Federal Highway Administration)

228

CHAPTER 9 GROUND PROPERTIES


values can be used as indicators of the degree of weathering and soundness of the rock, as well as compared with in-situ field measurements
that relate to the extent of fissuring and discontinuities of the larger
rock mass. The summary of data in Fig. 9.5 illustrates the general
ranges of compression wave (Vp ) between 3000 and 7000 m/s and
ranges of shear waves (Vs ) between 2000 and 3500 m/s for intact rocks.

Compressive strength
The stressstrain-strength behaviour of intact rock specimens can be
measured during a uniaxial compression test (unconfined compression),
or the more elaborate triaxial test. The peak stress of the stressstrain
curve during unconfined loading is the uniaxial compressive strength
(designated qu or u ). The value of qu can be estimated from the point
load index (Is ) that is easily conducted in the field. Representative
values of compressive strengths for a variety of intact rock specimens
are listed in Table 9.4. For this database, the compressive strengths
Table 9.4 Representative measured parameters on intact rock specimens (after Goodman,
1989 from Mayne et al., 2002; permission John Wiley and Sons, Inc.)
Intact rock material

qu :
MPa

T0 :
MPa

Baraboo quartzite
Bedford limestone
Berea sandstone
Cedar City tonalite
Cherokee marble
Dworshak Dam gneiss
Flaming Gorge shale
Hackensack siltstone
John Day basalt
Lockport dolomite
Micaceous shale
Navajo sandstone
Nevada basalt
Nevada granite
Nevada tuff
Oneota dolomite
Palisades diabase
Pikes Peak granite
Quartz mica schist
Solenhofen limestone
Taconic marble
Tavernalle limestone

320.0
51.0
73.8
101.5
66.9
162.0
35.2
122.7
355.0
90.3
75.2
214.0
148.0
141.1
11.3
86.9
241.0
226.0
55.2
245.0
62.0
97.9

11.0
1.6
1.2
6.4
1.8
6.9
0.2
3.0
14.5
3.0
2.1
8.1
13.1
11.7
1.1
4.4
11.4
11.9
0.5
4.0
1.2
3.9

Statistical results: Mean


S. Dev.

135.5
93.7

5.6
4.7

:


Ratio:
qu =T0

Ratio:
ER =qu

88 320
28 509
19 262
19 184
55 795
53 622
5526
29 571
83 780
51 020
11 130
39 162
34 928
73 795
3649.9
43 885
81 699
70 512
20 700
63 700
47 926
55 803

0.11
0.29
0.38
0.17
0.25
0.34
0.25
0.22
0.29
0.34
0.29
0.46
0.32
0.22
0.29
0.34
0.28
0.18
0.31
0.29
0.40
0.30

29.1
32.3
63.0
15.9
37.4
23.5
167.6
41.5
24.5
29.8
36.3
26.3
11.3
12.1
10.0
19.7
21.1
19.0
100.4
61.3
53.0
25.0

276
559
261
189
834
331
157
241
236
565
148
183
236
523
323
505
339
312
375
260
773
570

44 613
25 716

0.29
0.08

39.1
35.6

ER :
MPa

372.5
193.8

Note: 1 MPa 10.45 tsf 145.1 psi

229

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


1

10

Note: 1 MN/m2 = 145 psi

100

Very weak

Weak

Very low

Very weak
Soil

Low

Moderately
weak

Weak

Moderately
strong

Very strong

Medium High

Very high

Very
Strong strong

Extremely
strong

Coates
(1964)
Deere and Miller
(1966)
Geological Society
(1970)

Rock

Extremely
low

Very low

Soil

Very soft

Low

Medium

Soft

Soil

Hard

Low

Low

Medium

Jennings
(1973)

High

Bieniawski
(1973)

High

10

20

50

Very high

Very high

Very
strong

100

200

ISRM
(1979)
Kulhawy et al.
(1991)

STRONG

Medium Strong
strong

Weak

Extremely Broch and Franklin


high
(1972)

Extremely hard

MEDIUM

Very weak

Very high

Moderate Medium

WEAK
Extremely
weak

High

Very hard

Very low

Very low

0.5

Strong

Extremely BS EN
strong
14689-1: 2003

500

Uniaxial compressive strengh: MN/m2

Fig. 9.6 Classifications for unweathered intact rock material strength


(Kulhawy et al., 1991; from Mayne et al., 2002; permission ASCE)
ranged from 11 to 355 MPa, with a mean value of qu 135 MPa. A wide
range in compressive strength can exist for a particular rock type,
depending upon porosity, cementation, degree of weathering, formation
heterogeneity, grain size angularity and degree of interlocking of
mineral grains. The compressive strength also depends upon the orientation of load application with respect to microstructure, for example foliation in metamorphic rocks and bedding planes in sedimentary rocks.
The compressive strength serves as an initial index on the competency
of intact rock. Figure 9.6 shows a comparison of several classification
schemes. This is particularly useful for defining differences between
hard clays to shales, as the boundary in the transition from soil to rock is
not precise in these sedimentary materials. Similarly, it is applicable to
residual profiles where the transition from soil to saprolite and weathered
rock and rock may be needed. It can become important in contracts
involving excavatability issues of rock versus soil, as the former is
considerably more expensive than the latter during site grading, deep
excavations and foundation construction.

Direct and indirect tensile strength


Rock is relatively weak in tension, and thus the tensile strength (T0 ) of
an intact rock is considerably less than its compressive value (qu ). Their
230

CHAPTER 9 GROUND PROPERTIES


= normal stress

Triaxial
Uniaxial

= normal stress
t = T0

0 3

Tension

u = qu

Compression

Fig. 9.7 Interrelationships between uniaxial compression, triaxial and


tensile strengths of intact rock in the MohrCoulomb diagram (from
Mayne et al., 2002; permission Federal Highway Administration)
interrelation in terms of Mohr strength criterion is shown in Fig. 9.7. The
direct tensile strength on rock specimens is not a common laboratory
procedure because of the difficulties involved in proper end preparation
(Jaeger and Cook, 1977). Therefore, it is usual to evaluate the tensile
strength through indirect methods, including the split-tensile test (Brazilian test) or alternatively, a bending test to obtain the modulus of rupture.
A list of representative tensile strength values for various rocks is given
in Table 9.4 with a measured range from 0.2 to 14.5 MPa and mean value
T0 5:6 MPa. For the data considered, it can be seen from Fig. 9.8 that the
25
T0/qu = 0.04 0.01
Tensile strength, T0: MPa

20

15

Sedimentary
Metamorphic
Igneous
Trend
+ S.E.
S.E.

10

0
0

100

200

300

400

500

600

Compressive strength, qu: MPa

Fig. 9.8 Comparison of tensile versus compressive strengths for intact rock
specimens (from Mayne et al., 2002; permission Federal Highway
Administration)

231

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


tensile strength averages only about 4% of the compressive strength for
the same rock.

Elastic modulus of intact rock


The Youngs modulus (ER ) of intact rock is measured during uniaxial
compression or triaxial compression loading. The equivalent elastic
modulus is the slope of the stressstrain curve and can be assessed as
either a tangent value (E =") or a secant value (E =") from the
initial loading. Also, it may be evaluated from an unload-reload cycle
implemented off the initial loading ramp. Most common in engineering
practice, the tangent value taken at 50% of ultimate strength is reported
as the characteristic elastic modulus (ER50 ).
Intact rock specimens can exhibit a wide range of elastic moduli, as
evidenced by Table 9.4. For these data, the measured values vary from
3.6 to 88.3 GPa, with a mean value of ER 44:6 GPa. Notably, these
moduli are comparable to normal and high-strength concretes that are
manufactured for construction. For many sedimentary and foliated
metamorphic rocks, the modulus of elasticity is generally greater parallel
to the bedding or foliation planes than perpendicular to them, due to
closure of parallel weakness planes.
An intact rock classification system based on strength and modulus
ratio (E=u ) is given in Table 9.5. For each of the basic rock types
(igneous, sedimentary and metamorphic), Fig. 9.9 shows the corresponding
Table 9.5 Engineering classification of intact rock (Deere and Miller, 1966;
Stagg and Zienkiewicz, 1968; from Mayne et al., 2002; permission USAF,
Kirtland, NM)
I. On basis of strength, u
Class

Description

A
B
C
D
E

Very high strength


High strength
Medium strength
Low strength
Very low strength

Uniaxial compressive strength: MPa


Over 220
110220
55110
2855
Less than 28

II. On basis of modulus ratio, Et =u


Class

Description

H
M
L

High modulus ratio


Average (medium) ratio
Low modulus ratio

a
b

232

Modulus ratiob
Over 500
200500
Less than 200

Rocks are classified by strength and modulus ratio such as AM, BL, BH, CM, etc.
Modulus ratio Et =ault where Et is tangent modulus at 50% ultimate strength and ault is the
uniaxial compressive strength.

(kg/cm2 105)

CHAPTER 9 GROUND PROPERTIES

28
E
Very low
strength

16

56

112

224

(MPa)

D
C
B
A
Low
Medium
High Very high
strength strength strength strength

1 Diabase
2 Granite family
3 Basalt and other flow rocks

us

ra
tio

m
od
ul
io

us
ul

od

lo
w

0:
20

0.5

ra
t

av
er
ag

50
0:
1

us

hi
gh

ra
tio

m
od
ul

Youngs modulus, Et

0.25
75

125

250
500
1000
2000
Uniaxial compressive strength, a(ult)

4000 (kg/cm2)

(a)

Fig. 9.9 Elastic moduluscompressive strength groupings for intact rock


materials for (a) igneous rocks, (b) sedimentary rocks and (c) metamorphic
rocks (after Deere and Miller, 1966; from Mayne et al., 2002; permission
USAF, Kirtland, NM)
groupings of elastic modulus (Et ) versus uniaxial compressive strength (u ).
The modulus here is the tangent modulus at 50% of ultimate strength. The
broad range of strengths and moduli shown in the three figures is
informative. The above system considers intact rock specimens only and
does not consider the natural fractures (discontinuities) in the rock mass.
For lab testing on intact rock specimens, the nondestructive elastic modulus
at very small strains is obtained from ultrasonic measurements and this
value is higher than moduli measured at intermediate to high strains, such
as Et50 . Figure 9.10 shows a global database of Emax from small-strain
measurements (ultrasonics, bender elements, resonant column) versus the
compressive strength (qmax qu ) for a wide range of civil engineering
materials ranging from soils to rocks, as well as concrete and steel.

Operational shear strength


The shear strength of rock usually controls in the geotechnical evaluation
of slopes, tunnels, excavations and foundations. As such, the shear

233

(kg/cm2 105)

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS

28
E
Very low
strength

16

56

112

224

(MPa)

D
C
B
A
Low
Medium
High Very high
strength strength strength strength

1 Limestone and dolomite


2 Sandstone
3 Shale
8

us

ra
tio

us

ra
t

io

e
L

2
lo 00:
w 1
m
od
ul

av
er
ag

0.5

m
od
ul

50
0:
1

us

hi
gh

ra
tio

m
od
ul

Youngs modulus, Et

0.25
75

125

250
500
1000
2000
Uniaxial compressive strength, a(ult)

4000 (kg/cm2)

(b)

Fig. 9.9 continued


strength () of in-place rock often needs to be defined at three distinct
levels: (a) intact rock, (b) along a rock joint or discontinuity plane,
and (c) representative of an entire fractured rock mass. Figure 9.11
illustrates these cases for the illustrative example involving a road
highway cut in rock. In all cases, the shear strength is most commonly
determined in terms of the MohrCoulomb criterion (Fig. 9.7):
 c0 0 tan 0

7:3

where  operational shear strength, 0 effective normal stress on the


plane of shearing, c0 effective cohesion intercept and 0 effective
friction angle. The appropriate values of the MohrCoulomb parameters
c0 and 0 will depend greatly upon the specific cases considered and
levels of failure applicable as illustrated in Fig. 9.11.
For the intact rock, a series of triaxial compressive strength tests can be
performed at increasing confining stresses to define the MohrCoulomb
envelope and corresponding c0 and 0 parameters. Alternatively, empirical
methods based on the type of rock material and its measured uniaxial
compressive strength (qu u ) are available for evaluating the shear
strength parameters of intact rock (Hoek et al., 1995), as discussed later.

234

(kg/cm2 105)

CHAPTER 9 GROUND PROPERTIES

28
E
Very low
strength

16

1 Quartzite
2 Gneiss
3 Marble
4 Schist
a Steep foliation
b Flat foliation

112

224

(MPa)

1
3
2

us

ra
tio

4a

m
od
ul

ra
tio

4b

ra
t
us
ul
od

lo
w

0:
20

0.5

io

av
er
ag

m
od
ul

50
0:
1

us

hi
gh

Youngs modulus, Et

56

D
C
B
A
Low
Medium
High Very high
strength strength strength strength

0.25
75

125

250
500
1000
2000
Uniaxial compressive strength, a(ult)

4000 (kg/cm2)

(c)

Fig. 9.9 continued


This approach is versatile as it can be reduced to account for the degree of
fracturing and weathering, thus also used to represent and estimate the
shear strength of rock masses.
Laboratory direct shear testing can be used to determine the shear
strength of a discontinuity and/or the infilling material found within the
joints. The split box is orientated with the axis along the preferred plane
of interest. The shear strength of the discontinuity surface has either a
representative peak or residual value of the frictional component of
shear strength. Peak shear strengths will apply during highway cuts
and excavations in rocks where no movement has occurred before.
Residual shear strengths will be appropriate in restoration and remedial
work involving rockslides and slipped wedges or blocks of rock. Relatively
small movements can reduce shear strength from peak to residual
values. The peak values can be conceived as the composite of the residual
shear strength and a geometrical component that depends on roughness
and related to asperities and roughness on the joint plane. Table 9.6
lists values of peak friction angle of various rock surface types, rock
minerals (that may coat the joints) and infilling materials (such as clays

235

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


107
Emax at 1 < 105

Magnus clay (after Jardine, 1985)


London clay (after Jardine, 1985)
Sand (ML TSS, = 0.5 assumed)
Sand (static)
Gravel

Emax: kgf/cm2

106

105

Range for artificial materials


having cementation

104
Steel (SS41)
Worldwide data of
various rocks (dynamic)
Concrete
Cement treated sand
Sagamihara mudstone
Sagara mudstone
Laminated sandstone
and mudstone

103
Emax/qmax = 102

Emax/qmax = 103
102
101

101

10

102

103

104

qmax: kgf/cm2

Fig. 9.10 Small-strain elastic modulus (Emax ) versus compressive strength


(qu ) for all types of civil engineering materials (after Tatsuoka and
Shibuya, 1992; from Mayne et al., 2002; permission Prof. F. Tatsuoka)
and sands). If the joints are sufficiently open, the infilling of clay/soil may
dominate the shear strength behaviour of the situation.
Movement reduces (or removes) the effect of the asperities, resulting in
reduced shear strength. If sufficient movement occurs, the residual
strength of the material is reached. Table 9.7 presents a selection of
reported values of residual frictional angle (0r , assuming c0r 0) for
various types of rock surfaces and minerals found in rock joints and
discontinuities. These values can give an approximate guide in selecting
interface and joint strengths.
Assumed slip plane

Sloping joints

Sloping joints

Extensive cracks

Cut slope

Road

(a) Massive rock

Wedge

Cut slope

(c) Unfavourable joint set

(d) Highly fractured mass

Road

(b) Favourable joint set

Fig. 9.11 Illustrative cases for defining rock shear strength for a cutting in rock for the cases
of (a) intact rock, (b) intact strength across joints, (c) shear strength along joint planes and (d)
jointed rock mass (from Mayne et al., 2002; permission Federal Highway Administration)

236

CHAPTER 9 GROUND PROPERTIES


Table 9.6 Friction angles for rock joints, minerals and fillings (after
compilations by Franklin and Dusseault, 1989; Jaegar and Cook, 1977; from
Mayne et al., 2002; permission McGraw-Hill)
Condition/case
Thick joint fillings:
Smectite and montmorillonitic clays
Kaolinite
Illite
Chlorite
Quartzitic sand
Feldspathic sand
Minerals:
Talc
Serpentine
Biotite (mica)
Muscovite (mica)
Calcite
Feldspar
Quartz
Rock joints:
Crystalline limestone
Porous limestone
Chalk
Sandstone
Quartzite
Clay shale
Bentonitic shale
Granite
Dolerite
Schist
Marble
Gabbro
Gneiss

Friction angle 0 (degree): c0 0


510
1215
1622
2030
3340
2835
9
16
7
13
8
24
33
4249
3248
3041
2435
2344
2237
927
3133
3343
3240
3137
33
3135

Additional guidelines for the selection of MohrCoulomb parameters


are given by Hoek et al. (1995) and Wyllie (1999).

Rock mass classication


While the mineral composition, age and porosity determine the properties
of the intact rock, the network of fractures, cracks and joints governs the
rock mass behaviour in terms of available strength, stiffness, permeability
and performance. The pattern of discontinuities of the rock mass will
be evident in the cored sections obtained during the site exploration
studies, as well as in the exposed faces and rock outcrops in the
topographic terrain. A selection of exposed rock types is presented in

237

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


Table 9.7 Residual friction angles (compilations after Barton, 1973; Hoek
and Bray, 1977; from Mayne et al., 2002; permission Elsevier)
Rock type

Residual friction angle r (degrees),


assuming c0 0

Amphibolite
Basalt
Conglomerate
Chalk
Dolomite
Gneiss (schistose)
Granite (fine grain)
Limestone
Porphyry
Sandstone
Shale
Siltstone
Slate

32
3138
35
30
2731
2329
2935
3340
31
2535
27
2731
2530

Note: Lower value is generally given by tests on wet rock surfaces.

Fig. 9.12 to illustrate the variations that occur in scenery due to the
inherent fracture and joint patterns.
Means for quantifying the degree, extent and nature of the discontinuities is paramount in assessing the quality and condition of the rock mass.
The rock quality designation (RQD) is a first-order assessment of the
amount of natural jointing and fissuring in rock masses. The RQD has
been used to approximately quantify the rock mass behaviour, yet was
developed about four decades ago (Deere and Deere, 1989). Since then,
more elaborate and quantitative methods of assessing the overall rock
mass condition have been developed, including the Geomechanics
RMR-System (Bieniawski, 1989), based on mining experiences in South
Africa, and the NGI-Q system (Barton, 1988), based on tunnelling experiences in Norway. A closely related system to the RMR is the Geological
Strength Index (GSI) that is useful in assessing the strength of rock
masses. These and other rock mass classifications systems are summarised in ASTM D 5878 (Classification of Rock Mass Systems). The
influential factors that comprise the rock mass ratings will be briefly
discussed here and presented in the context for the interpretation of
rock mass properties needed for design and analysis of slopes, tunnels,
and foundations in rock formations.

Rock Mass Rating System (RMR)


The Rock Mass Rating (RMR) rock classification system uses five basic
parameters for classification and properties evaluation. A sixth parameter
helps further assess issues of stability to specific problems. Originally

238

CHAPTER 9 GROUND PROPERTIES

Fig. 9.12 Selection of exposed rock masses from different geologic origins (from Mayne
et al., 2002; permission Federal Highway Administration)
(a) Limestone at Interstate Highway 75, Tennessee, USA
(b) Sandstone in Grand Canyon, Arizona, USA
(c) Basalt beach, Kauai, Hawaii, USA
(d) Mica schist near Hope, British Columbia, Canada
(e) Gneiss at Sondestrom, Greenland
(f ) Exposed granite, Rio de Janeiro, Brazil.

239

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


intended for tunnelling and mining applications, it has been extended for
the design of cut slopes and foundations. The six parameters used to
determine the RMR value are:
(i)
(ii)
(iii)
(iv)
(v)
(vi)


uniaxial compressive strength (qu or u )


Rock Quality Designation (RQD)
spacing of discontinuities
condition of discontinuities
groundwater conditions
orientation of discontinuities.

Note: Value may be estimated from point load index (Is ).

The basic components of the RMR system are contained in Fig. 9.13. The
rating is obtained by summing the values assigned for the first five
components. Later, an overall rating can be made by a final adjustment
by consideration of the sixth component, depending upon the intended
project type (tunnel, slope or foundation); however, this is less utilised
in most routine applications. Thus, the RMR is determined as:
RMR

5
X

Ri

7:4

i1

The RMR rating assigns a value of between 0 (very poor) to 100 (most
excellent) for the rock mass. The RMR system has been modified over
the years with additional details and variants given elsewhere (Bieniawski, 1989; Hoek et al., 1995; Wyllie, 1999). Depending upon the dip
and dip direction (or strike) of the natural discontinuities with respect to
the proposed layout and orientation of the construction, then an additional
factor may be added to adjust the RMR, ranging from favourable (R6 0)
to very unfavourable (12 for tunnels, 25 for foundations and 60 for
slopes).

NGI Q Rating
The Q Rating was developed for assessing rock masses for tunnelling
applications by the Norwegian Geotechnical Institute (Barton et al.,
1974) and relies on six parameters for evaluation:
(i)
(ii)
(iii)
(iv)
(v)

240

Rock Quality Designation (RQD)


Jn is the number of discontinuity sets in the rock mass (joint sets)
Jr represents the roughness of the interface within the discontinuities, fractures and joints
Ja describes the condition, alterations and infilling material with the
joints and cracks
Jw provides assessment on the in-place water conditions

CHAPTER 9 GROUND PROPERTIES


Rock Mass Rating (RMR)

also CSIR System

Geomechanics System (Bieniawski, 1984, 1989)


Geomechanics classification for rock masses
Class Description
Range of RMR
I
Very good rock
81 to 100
II
Good rock
61 to 80
III
Fair rock
41 to 60
IV
Poor rock
21 to 40
V
Very poor rock
0 to 20

NOTE: Rock Mass Rating is obtained by summing the five index


parameters to obtain an overall RMR. Adjustments for dip and
orientation of discontinuities being favourable or unfavourable
for specific cases of tunnels, slopes and foundations can also
be considered.
25

16

RMR rating, R2

RMR rating, R1

14
12
10
8
6
4
2
0

20
15
10
5
0

50

100

150

200

250

300

10

20

Unconfined compressive strength, qu: MPa

30

40

50

60

70

80

90

100

Rock Quality Designation, RQD


35

25

RMR rating, R4

RMR rating, R3

Rough/unweathered

30

20
15
10
5

Slightly

25

rough

Weathered

20
15

Slickensided surface or gouge-filled

10
5

0
0.01

Soft gouge-filled

0
0.1

10

Joint spacing: m
16

16
u = joint water pressure
1 = major principal stress

14
12
10

RMR rating, R6

RMR rating, R5

Joint separation or gouge thickness: mm

Alternate 2 definition
for parameter R5

8
6

Alternate 2 definition
for parameter R5

12
Damp

10

Wet

8
6

Dry

14

Dripping

Flowing

0
0

0.1

0.2

0.3

0.4

0.5

Joint water pressure ratio, u/1

0.6

10

100

1000

Inflow per 10 m tunnel length: litres/min

Fig. 9.13 The geomechanics classification system for Rock Mass Rating (RMR) (after
Bieniawski, 1984, 1989; from Mayne et al., 2002; permission Taylor & Francis)
(vi)

SRF is a stress reduction factor related to the initial stress state and
compactness.

The individual parameters are assigned values as per the criteria given in
Fig. 9.14 and then a complete Q rating is obtained as follows:


 
RQD
Jr
Jw
Q
7:5
Jn
Ja
SRF

241

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS

Norwegian classif ication f or rock masses


Q value
<0.01
0.01
0.1
1
4
10
40
100
<400

to 0.1
to 1
to 4
to 10
to 40
to 100
to 400

Q = (RQD/Jn)(Jr/Ja)(Jw/SRF)

Quality of rock mass


4. Discontinuity condition and infilling = Ja
4.1 Unfilled cases

Exceptionally poor
Extremely poor
Very poor
Poor
Fair
Good
Very good
Extremely good
Exceptionally good

Healed
Stained, no alteration
Silty or sandy coating
Clay coating

0.75
1
3
4

1. RQD = Rock Quality Designation = sum of cored pieces


>100 mm long, divided by total core run length

4.2 Filled discontinuities


Sand or crushed rock infill
Stiff clay infilling <5 mm
Soft clay infill <5 mm thick
Swelling clay <5 mm
Stiff clay infill >5 mm thick
Soft clay infill >5 mm thick
Swelling clay >5 mm

2. Number of sets of discontinuities (joint sets) = Jn


Massive
One set
Two sets
Three sets
Four or more sets
Crushed rock

5. Water conditions
Dry
1
Medium water inflow
0.66
Large inflow in unfilled joints
0.5
Large inflow with filled joints that wash out 0.33
High transient flow
0.2 to 0.1
High continuous flow
0.1 to 0.05

Parameters for the Q-rating of rock masses

0.5
2
4
9
15
20

3. Roughness of discontinuities* = Jr
Non-continuous joints
Rough, wavy
Smooth, wavy
Rough, planar
Smooth, planar
Slick and planar
Filled discontinuities
*Note: add +1 if mean joint spacing >3 m

4
6
8
12
10
15
20

6. Stress reduction factor** = SRF


4
3
2
1.5
1
0.5
1

Loose rock with clay infill


Loose rock with open joints
Shallow rock with clay infill
Rock with unfilled joints

10
5
2.5
1

**Note: Additional SRF values given for rocks prone to


bursting, squeezing and swelling by Barton et al. (1974)

Fig. 9.14 The Q-Rating system for Rock Mass Classification (after Barton et al., 1974;
from Mayne et al., 2002; permission Elsevier)
Both the RMR and the Q ratings can be used to evaluate the stand-up
time of unsupported mine and tunnel walls which is valuable during
construction. The RMR and Q are also used to determine the type and
degree of tunnel support system required for long-term stability, including
the use of shotcrete, mesh, lining and rock bolt spacing. Details on these
facets are given elsewhere (Hoek et al., 1995).

Geological Strength Index (GSI)


Whereas the RMR and Q systems were developed originally for mining
and tunnelling applications, the Geological Strength Index (GSI) provides
a measure of the rock mass quality for directly assessing the strength and
stiffness of intact and fractured rocks. A quick assessment of the GSI may
be made by use of the graphical chart given in Fig. 9.15, thus facilitating
the procedure for field use.

242

CHAPTER 9 GROUND PROPERTIES

Structure

Very poor
Slickensided, highly weathered
surfaces with soft clay coatings or
fillings

Poor
Slickensided or highly weathered
surfaces with compact coatings or
fillings of angular fragments

Fair
Smooth, moderately weathered and
altered surfaces

Good
Rough, slightly weathered, iron
stained surfaces

GSI Value

Surface conditions

From the description of structure and surface conditions of


the rock mass, pick an appropriate box in this chart. Estimate
the average value to the Geological Strength Index (GSI)
from the contours. Do not attempt to be too precise. Quoting
a range of GSI from 36 to 42 is more realistic than stating
GSI = 38. It is also important to recognise that the
HoekBrown criterion should only be applied to rock
masses where the size of individual blocks are small
compared with the size of the excavation under
consideration.

Very good
Very rough and fresh unweathered
surfaces

Geological Strength Index

Decreasing surface quality

Blocky very well interlocked undisturbed rock


mass consisting of cubical blocks formed by
three orthogonal discontinuity sets

80

70

60

Very blocky interlocked, partially disturbed rock


mass with multifaceted angular blocks formed by
four or more discontinuity sets

50

Disintegrated poorly interlocked, heavily


broken rock mass with a mixture of angular
and rounded rock pieces

Foliated/laminated/sheared thinly laminated or


foliated, tectonically sheared weak rocks,
closely spaced schar??? prevails over any
other discontinuity set, resulting in complete
lack of blockiness

40

Decreasing interblocking of rock pieces

Blocky/disturbed folded and/or faulted with


angular blocks formed by many intersecting
discontinuity sets

30

20

10

N/A

N/A

Fig. 9.15 Chart for estimating the Geological Strength Index (GSI) (after Hoek and
Brown, 1997; from Mayne et al., 2002; permission Elsevier)

243

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


More specifically, the GSI can be calculated from the components of the
Q system, as follows:
 

RQD
Jr
44
7:6
GSI 9 log
Jn
Ja
In relation to the common Geomechanics Classification System, the GSI is
restricted to RMR values in excess of 25, thus:
for RMR > 25: GSI

4
X

Ri 10

7:7

i1

Rock mass strength


The strength of the overall assemblage of rock blocks and fractures can
be assessed by large direct shear tests conducted in the field, backcalculation of rockslides and failured slopes, or alternatively estimated
on the basis of rock mass classification schemes. For the latter, a detailed
approach to evaluating the rock mass strength is afforded through use
of the GSI rating (Hoek et al., 1995). In this method, the major principal
stress (01 ) is related to the minor principal stress (03 ) at failure through
an empirical expression that depends upon the following:
(i)
(ii)
(iii)

The uniaxial compressive strength of the rock material (u )


A material constant (mi ) for the type of rock
Three empirical parameters that reflect the degree of fracturing of
the rock mass (mb , s and a)

The relationship accounts for curvature of the MohrCoulomb strength


envelope and gives the expression for major principal stress in the form:

a
0
01 03 u mb 3 s
7:8
u
The material parameter mi depends on the specific rock type (igneous,
metamorphic or sedimentary) as determined from the chart given in
Fig. 9.16. Values range as low as 4 for mudstone to as high as 33 for
gneiss and granite.
For GSI > 25, the remaining strength parameters for undisturbed rock
masses are:
mb mi expGSI  100=28

7:9

s expGSI  100=9

7:10

a 0:5

7:11

For GSI < 25, the parameter selection is given by:

244

s0

7:12

a 0:65  GSI=200

7:13

CHAPTER 9 GROUND PROPERTIES

Rock type

Class

Group

Clastic

Texture
Course

Medium

Fine

Very fine

Conglomerate
(22)

Sandstone
19

Siltstone
9

Claystone
4

Greywacke
(18)
Chalk
7

Sedimentary

Organic

Non-clastic

Carbonate

Coal
(821)
Breccia
(20)

Metamorphic

Chemical

Micritic
limestone
8

Gypstone
16

Anhydrite
13

Non-foliated

Marble
9

Hornfels
(19)

Quartzite
24

Slightly foliated

Migmatite
(30)

Amphibolite
31

Mylonites
(6)

Foliated

Gneiss
33

Schists
(10)

Phyllites
(10)

Slate
9

Granite
33

Rhyolite
(16)

Obsidian
(19)

Granodiorite
(30)

Dacite
(17)

Diorite
(28)

Andesite
19

Light

Igneous

Sparitic
limestone
(10)

Dark

Gabbro
27

Dolerite
(19)

Basalt
(17)

Breccia
(18)

Tuff
(15)

Norite
22
Extrusive pyroclastic type

Agglomerate
(20)

Fig. 9.16 Material constant mi for GSI evaluation of rock mass strength
(after Hoek et al., 1995; from Mayne et al., 2002; permission Taylor & Francis)
Thus, the evaluation is easily carried out using a spreadsheet with
adopted values of effective confining stresses (03 ) taken over the range
of anticipated field overburden stresses to calculate corresponding
values of effective major principal stress at failure (01 ) by equation (7.8).
Then, the paired values of 01 and 03 can be plotted (using either Mohr
circles or qp plots) to obtain the equivalent shear strength parameters,
c0 and 0 . Note that the method can also be applied to evaluate the strength
of intact rock (GSI 100), as well as fractured rock. For quick assessments, representative and average values of 03 have been used to
derive approximate chart solutions for selecting normalised c0 =u and
friction angle 0 directly from GSI and material constant mi , as presented
in Fig. 9.17.

245

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


0.20

mi

55

35
30
25
20
16
13

0.08
0.06
0.05
0.04
0.03

mi
35
30
25
20
16
13
10

0.02

40

10

35

30

25
20
15

7
5

10

45
Friction angle: degrees

0.10

Cohesive strength/uniaxial strength

50

0.01
20

30

40

50

60

70

80

Geological Strength Index, GSI

0.008
90

10
10

20

30

40

50

60

70

80

90

Geological Strength Index, GSI

Fig. 9.17 Approximate chart solution for obtaining normalised cohesion intercept
(c0 =u ) and friction angle 0 from GSI rating and mi parameter (after Hoek and Brown,
1997; from Mayne et al., 2002; permission Elsevier)
For the apparent shear strength along specific joints and planes of
sliding, the peak friction angle can be evaluated from the Q rating
parameters (c0 0):
 
Jr
7:14
0p 
Ja
which gives a range of 78 < 0p < 758 for the full value limits of joint
roughness (Jr ) and alteration (Ja ) parameters.

Rock mass modulus


The equivalent elastic modulus (EM ) of rock masses is used in deformation
analyses and numerical simulations involving tunnels, slopes and foundations, to estimate magnitudes of movements and deflections caused by
new loading. Field methods of measuring the deformability characteristics
of rock masses include the Goodman jack and rock dilatometer, as well as
back-calculation from full-scale foundation load tests (Littlechild et al.,
2000). For routine calculations, EM has been empirically related to intact
rock properties (uniaxial strength, u , and elastic modulus of the intact
rock, ER ), rock quality (RQD) and rock mass ratings (RMR, Q and GSI),
such as given by the expressions listed in Table 9.8. On critical projects,
the actual stiffness of the rock formation can be assessed using full-scale
load tests, made more practical in recent times by the advent of the

246

CHAPTER 9 GROUND PROPERTIES


Table 9.8 Empirical methods for evaluating elastic modulus (EM ) of rock masses (from
Mayne et al., 2002; permission Federal Highway Administration)
Expression

Notes/remarks

Reference

For RQD < 70: EM ER (RQD/350)


For RQD > 70: EM ER [0:2 RQD  7037:5]

Reduction factor on
intact rock modulus

Bieniawski (1978)

EM  ER 0:1 RMR=1150  11:4RMR

Reduction factor

Kulhawy (1978)

EM GPa 2RMR  100

45 < RMR < 90

Bieniawski (1984)

EM GPa 25 log10 Q

1 < Q < 400

Hoek et al. (1995)

EM GPa 10

0 < RMR < 90

Serafim and
Pereira (1983)

EM GPa 0:01u 10GSI100=40

Adjustment for rocks


with u < 100 MPa

Hoek (1999)

RMR100=40

Notes: ER intact rock modulus, EM equivalent rock mass modulus, RQD rock quality designation, RMR rock
mass rating, Q NGI rating of rock mass, GSI geologic strength index, u = uniaxial compressive strength.

Osterberg load cell which can apply very large forces using embedded
hydraulic systems.

Foundation resistances
In many construction projects, foundations can bear on the rock surface or
be embedded into the rock formation to resist large axial loads. For bridge
structures, for example, shallow spread footing foundations not subjected
to scour can bear directly on the rock. In other instances, deep foundations
may consist of large drilled shafts or piers that are constructed into the
rock using coring methods. These may be designed for axial compression
and/or uplift. In the following sections, methods of estimating the bearing
stresses and side resistance in rocks are provided.
Allowable foundation bearing stress
Detailed calculations can be made concerning the bearing capacity of
foundations situated on fractured rock (Goodman, 1989). In addition, the
results of the field and laboratory characterisation programme of the
rock mass may be used to estimate the allowable bearing values directly.
In the most simple approach, presumptive values are obtained from local
practice, Uniform and BOCA building codes and AASHTO guidelines. A
summary of allowable bearing stresses from codes has been compiled by
Wyllie (1999) and presented in Fig. 9.18. For RQD < 90%, the values given
in the figure should be decreased by variable reduction factors ranging
from 0.7 to 0.1. In this regard, the approach of Peck et al. (1974) uses the
RQD directly to assess the allowable bearing stress (qallowable ), provided
that the applied stress does not exceed the uniaxial compressive strength

247

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


Rock type

Age

Locations
1

Granite
Manhattan schist
Fordham gneiss
Dolomite
Massively bedded limestone

Ontario
New York
New York
Chicago
UK

Pre. Camb
Pre. Camb
L. Paleoz

Mica schist
Limestone
Hard, cemented shale
Austin chalk
Dolomite
Clay shale
Pierre shale
Fox Hills sandstone
Hard, very dense glacial till
Eagleford shale
Solid chalk
Limestone
Mica schist
Schist and slate
Argillite
Newark shale
Friable sandstone
Friable claystone and sandstone

Allowable bearing pressure: MPa


2 3 4 5 6 7 8 9 10

Pre. Camb
U. Paleoz

Philadelphia
St Louis
UK
Dallas
Detroit
UK
Denver
Denver
Ontario
Dallas
UK
Kansas City
Washinton
UK
Cambridge MA
Philadelphia
Los Angeles
Oakland CA

Cretaceous
L. Paleoz
Cretaceous
Tertiary
Cretaceous
Cretaceous
U. Paleoz
Pre. Camb
Pre. Camb
Triassic
Quarternary
Tertiary

20 40 60 80 100 120 140 160 180 200 220


kips/ft2

Fig. 9.18 Allowable bearing stresses on unweathered rock from codes of practice (after
Wyllie, 1999; from Mayne et al., 2002; permission Cengage)
Foundations on fr actur ed rock for mations

30

Allowable bearing stress qa: MPa

Note: Use maximum qa < qu


where qu = compressive strength of intact rock specimens

25

q allowable (MPa) ? 1 +

20

(RQD/16)
1 (RQD/130)

Note: 1 MPa = 10 tsf

15

10

5
Peck et al. (1974)
Approximation
0
0

10

20

30

40
50
60
70
Rock quality designation, RQD

80

90

100

Fig. 9.19 Allowable bearing stresses on fractured rock from Rock Quality
Designation (RQD) (after Peck et al., 1974; from Mayne et al., 2002;
permission John Wiley and Dr Ralph B. Peck)

248

CHAPTER 9 GROUND PROPERTIES


Table 9.9 Approximate allowable bearing values (after BS 8004, 1986; permission British
Standards Institution)
Group

Types of rocks and soils

Approximate
Remarks
bearing value: kPa

1 Rocks

Hard igneous and gneissic rocks in


sound condition
Hard limestones and hard sandstones
Schists and slates
Hard shales, hard mudstones and soft
sandstones
Soft shales and soft mudstones
Hard sound chalk, soft limestone
Thinly bedded limestones,
sandstones, shales
Heavily shattered rocks

2 Noncohesive
soils

Compact gravel, or compact sand and


gravel
Medium dense gravel, or medium
dense sand and gravel
Loose gravel, or loose sand and gravel
Compact sand
Medium dense sand
Loose sand

3 Cohesive Very stiff boulder clays and hard


soils
clays
Stiff clays
Firm clays
Soft clays and silts
Very soft clays and silts

10 000
4000
3000
2000

These values are based on the


assumption that the foundations
are carried down to
unweathered rock

600 to 1000
600
To be assessed
after inspection
>600
200 to 600
<200
>300
100 to 300
<100
300 to 600

Width of foundation (B) not less


than 1 m. Groundwater level
assumed to be a depth not less
than B below the base of the
foundation

Group 3 is susceptible to longterm consolidation settlement

150 to 300
75 to 150
<75
Not applicable

of the intact rock (qallowable < u ). The RQD relationship with allowable
bearing stress is shown in Fig. 9.19. For more specific calculations and
detailed evaluations, the results of the equivalent MohrCoulomb parameters from the GSI approach may be used in traditional bearing capacity
equations, as discussed by Wyllie (1999). Some typical bearing values for
soils and rocks are given in Table 9.9.

Foundation side resistances


Deep foundations can be constructed to bear within rock formations to
avert scour problems and resist both axial compression and uplift
loading. Drilled shaft foundations can be bored through soil layers and
extended deeper by coring into the underlying bedrock. In many cases,
the diameter of the drilled shaft is reduced when penetrating the rock,
thus making a socket. Figure 9.20 presents a relationship between the
shaft side resistance ( fs ) and one-half the compressive strength (qu =2)
for sedimentary rocks, while Fig. 9.21 shows a similar diagram between
fs and qu for all rock types.
249

SHORT COURSE IN GEOLOGY FOR CIVIL ENGINEERS


100

Unit side resistance, fs: tsf

All load tests

10

Clay
Shale, mudstone
??? (rough ????)
Sandstone, limestone, marl
Rock

Clay
0.1
0.1

Soft

Med.

Stiff

V. stiff

Weak

Ext. stiff

10

Medium

Strong

100

1000

One-half compressive strength, qu/2: tsf

Fig. 9.20 Unit side resistance trend with strength of sedimentary rocks (after
Kulhawy and Phoon, 1993; from Mayne et al., 2002; permission of ASCE)

6000
Granitic

Volcanic

5000

Metasedimentary

Side resistance, fs: kPa

Sedimentary
Zhang and Einstein
rough socket (1998)

Rowe and Armitage


rough socket (1987)

4000

Rowe and Armitage


smooth socket (1987)
3000
Zhang and Einstein
smooth socket (1998)
2000
Horvath et al. (1983),
b = 0.3
Horvath et al. (1983)

1000

10

20

30

40

50

60

Uniaxial compressive strength, qu: MPa

Fig. 9.21 Maximum achieved side resistance versus unconfined compression strength
for various rock types (after Ng et al., 2001; permission of ASCE)

250

CHAPTER 9 GROUND PROPERTIES


Additional rock mass parameters
As projects become more complex, there is a need to measure and interpret additional geomechanical properties of the intact rock and rock
mass. Some recent efforts have included assessments of scour and erodibility that have been related to rock mass indices (Van Schalkwyk et al.,
1995). Similar methodologies have been developed for excavatability of
rocks by machinery in order to minimise use of blasting (Wyllie, 1999).
A simple approach for the latter purpose utilises the compression wave
velocity (VP ) of the in-place rock directly (Fig. 9.22).
Velocity: km
0

Topsoil
Clay
Glacial till
Igneous rocks
Granite
Basalt
Trap rock
Sedimentary rocks
Shale
Sandstone
Siltstone
Claystone
Conglomerate
Breccia
Callcite
Limestone
Metamorphic rocks
Schist
Slate
Minerals and ores
Coal
Iron ore
Rippable

Marginal

Non rippable

Fig. 9.22 Rippability of in-place rock by caterpillar dozer evaluated by


P-wave velocity (after Franklin and Dusseault, 1989; from Mayne et al.,
2002; permission of McGraw-Hill)

251