AFTES

GUIDELINES FOR

CARACTERISATION OF ROCK MASSES

USEFUL FOR THE DESIGN AND THE CONSTRUCTION
OF UNDERGROUND STRUCTURES
A.F.T.E.S. welcomes comments on this paper

Version 1 – Approved by technical Committe 29/04/2003

Draft submitted by
Jean-Louis GIAFFERI (EDF) - Chairman GT 1

With assistance in drafting this document from

Alain AMELOT (SPIE Batignolles) - Daniel ANDRE (SNCF) - François BERBET (BOUYGUES TP)
Philippe BOUSQUET-JACQ (EURISK) - Stéphane CURTIL (TERRASOL) - Jean-Louis DURVILLE (CETE Rhône-Alpes)
Denis FABRE (CNAM) - Jean-Alain FLEURISSON (Ecole des Mines de Paris - CGI) - Bernard GAUDIN (SCETAUROUTE)
Mehdi GOREYCHI (NERIS) - Françoise HOMAND (ENSG Nancy) - Gilles PARADIS (SNCF)
Jean PIRAUD (ANTEA) - Alain ROBERT (CETU) - Philippe VASKOU (GEOSTOCK)
Christophe VIBERT (Coyne et Bellier) - Françis WOJTKOWIAK (INERIS)
Special thanks are due to Bernard GAUDIN for shouldering the heavy burden of writing the many drafts needed before arriving
at this final report.
Thanks are also due to the proof-readers
Pierre DUFFAUT and Bernard GODINOT (GTM), Jean LAUNAY (VINCI Construction), Yann LEBLAIS (EEG SIMECSOL)

PREFACE
In 1978, AFTES issued its first Recommendations on the description of rock masses, using the following approach:
• Describe in detail all factors potentially influencing the stability of underground structures
• Classify field conditions with respect to each individual factor separately, without attempting to link them together.
This new version retains this basic principle, but with important additions.
Firstly, we make a clear distinction between the characterisation of (i) the rock matrix, (ii) discontinuities, and (iii) the rock mass
taken in its entirety, dealt with in the three main chapters forming the backbone of the Recommendations.
We have placed the description of the influencing factors within the general context of the geotechnical survey. This description is
dependent on a geological model. This model is made up of ‘homogeneous sub-units’ displaying the relevant characteristics. We
were also concerned about the transition from instrumental and field data to data values used in the design analyses.
Lastly, we took the decision – and risk – of presenting general rock classification systems, not in order to advocate their systematic
use but to point out their limitations. Despite their apparent convenience, these classification systems (and more importantly, the
correlations drawn from them) simplify to an outrageous degree a reality which is always complex. They can never be a substitute
for abundant observation, measurement and testing, and one must always bear in mind the value of the parameters on which
they are based throughout the whole design and construction process.
The descriptive approach recommended here by AFTES applies not only to structural stability analysis but equally to the selec-
tion of location, cross section and construction method. It is not confined to tunnels only, and these Recommendations may be
thought useful for other types of rock engineering.
Jean Piraud
Chairman, AFTES Technical Committee

2 T U N N E L S E T O U V R A G E S S O U T E R R A I N S - N ° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

CONTENTS

Pages Pages

1 - INTRODUCTION - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 4 4.1.2 - Degree of alteration - - - - - - - - - - - - - - - - - - - - - - 19

1.1 - PURPOSE OF ROCK MASS CHARACTERISATION - - - 4 4.1.3 - Rock mass continuity index ICM - - - - - - - - - - - - - 19
1.2 - GEOLOGICAL MODEL- - - - - - - - - - - - - - - - - - - - - - - - 4 4.2 - MECHANICAL PARAMETERS - - - - - - - - - - - - - - - - - - - 20
1.2.1 - Geological survey - - - - - - - - - - - - - - - - - - - - - - - - 4 4.2.1 - Rock mass deformability, rock mass deformation
1.2.2 - Geological model - - - - - - - - - - - - - - - - - - - - - - - - 4 modulus EMas - - - - - - - - - - - - - - - - - - - - - - - - - - 20
1.3 - GEOTECHNICAL CHARACTERISATION OF SUB-UNITS 4 4.2.2 - Rock mass limit strength - - - - - - - - - - - - - - - - - - - 21
2 - MATRIX CHARACTERISTICS- - - - - - - - - - - - - - - - - - - - - 5 4.3 - HYDROGEOLOGICAL CONDITIONS - - - - - - - - - - - - - 22
2.1 - IDENTIFICATION PARAMETERS- - - - - - - - - - - - - - - - - 6 4.3.1 - Identification of aquifers - - - - - - - - - - - - - - - - - - - 22
2.1.1 - Common names - - - - - - - - - - - - - - - - - - - - - - - - - 6 4.3.2 - Measurement of initial piezometric conditions - - - 22
2.1.2 - Petrography and mineralogy - - - - - - - - - - - - - - - - 6 4.3.3 - Measurement of rock mass permeability KM - - - - 22
2.1.3 - Alteration of the minerals in the rock matrix - - - - - 6 4.3.4 - Gas - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 23
2.1.4 - Densities (French standard P 94-410-1/2/3)- - - - - 6 4.3.5 - Other parameters - - - - - - - - - - - - - - - - - - - - - - - - 23
2.1.5 - Volume weights - - - - - - - - - - - - - - - - - - - - - - - - - 7 4.4 - INITIAL STATE OF STRESS IN ROCK MASS - - - - - - - - - 23
2.1.6 - Moisture content (French standard P 94-410-1)- - 7 4.4.1 - Initial state of stress and approximations - - - - - - - 23
2.1.7. Porosity (French standard P 94-410-3) - - - - - - - - - 7 4.4.2 - Characterisation of stress tensor - - - - - - - - - - - - - 24
2.1.8 - Degree of saturation - - - - - - - - - - - - - - - - - - - - - - 7 4.4.3 - Commentary on field test methods - - - - - - - - - - - 24
2.1.9 - Permeability - - - - - - - - - - - - - - - - - - - - - - - - - - - - 7 4.4.4 - Classification of stress states - - - - - - - - - - - - - - - - 25
2.1.10 - Ultrasound wave velocity 4.5 - TEMPERATURE - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 25
(French standard p 94-411) - Continuity index - - - - - - - - 7 4.5.1 - Geothermal parameters - - - - - - - - - - - - - - - - - - - 25
2.2 - MECHANICAL PARAMETERS - - - - - - - - - - - - - - - - - - - 7 4.5.2 - Methods for estimating temperatures for
2.2.1 - Deformability: instantaneous behaviour- - - - - - - - 8 underground structures - - - - - - - - - - - - - - - - - - - 25
2.2.2 - Deformability: time-dependent behaviour 5 - USE OF ROCK MASS CHARACTERISATION FOR
related to creep - - - - - - - - - - - - - - - - - - - - - - - - - 8 UNDERGROUND STRUCTURE STABILITY ANALYSIS
2.2.3 - Time-dependent behaviour related to swelling - - 9 AND CONSTRUCTION - - - - - - - - - - - - - - - - - - - - - - - - - - - 25
2.2.4 - Mechanical strength - - - - - - - - - - - - - - - - - - - - - - 9 5.1 - CHARATERISTIC VALUES AND PARAMETERS
2.2.5 - Triaxial test and failure criteria - - - - - - - - - - - - - - - 10 FOR DESIGN - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 25
2.2.6 - Parameters for resistance to excavation - - - - - - - - 11 5.1.1 - Individualisation of sub-units - - - - - - - - - - - - - - - - 25
2.2.7 - Other tests - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 12 5.1.2 - Geotechnical characterisation of sub-units - - - - - - 26
3 - CHARACTERISTICS OF DISCONTINUITIES - - - - - - - - - 12 5.2 - GEOTECHNICAL CLASSIFICATIONS - - - - - - - - - - - - - 27
3.1 - JOINT IDENTIFICATION PARAMETERS - - - - - - - - - - - 12 5.2.1 - General - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 27
3.1.1 - Types and origins of the discontinuities - - - - - - - - 12 5.2.2 - Bieniawski’s Rock Mass Rating - - - - - - - - - - - - - - - 27
3.1.2 -Description of discontinuities - - - - - - - - - - - - - - - - 13 5.2.3 - Barton’s Q index - - - - - - - - - - - - - - - - - - - - - - - - - 28
3.2 - CHARACTERISATION OF JOINT SYSTEMS- - - - - - - - - 13 5.2.4 - Summary and precautions - - - - - - - - - - - - - - - - - - 28
3.2.1 - Directional joint set patterns - - - - - - - - - - - - - - - - 13 5.3 - CORRELATIONS - - - - - - - - - - - - - - - - - - - - - - - - - - - - 29
3.2.2 - Statistical analysis of geometrical parameters 5.3.1 - General - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 29
for each joint set- - - - - - - - - - - - - - - - - - - - - - - - - 14 5.3.2 - Estimating rock mass deformability - - - - - - - - - - - 29
3.2.3 - Overall joint density indexes - - - - - - - - - - - - - - - - 15 5.3.3 - Hoek’s GSI index- - - - - - - - - - - - - - - - - - - - - - - - - 29
3.3 - MECHANICAL PARAMETERS OF DISCONTINUITIES - 17 5.3.4 - Estimating rock mass limit strength - - - - - - - - - - - 30
3.3.1 - Deformation parameters- - - - - - - - - - - - - - - - - - - 17 5.4 - PRESENTATION OF ROCK MASS
3.3.2 - Shear strength parameters - - - - - - - - - - - - - - - - - 17 CHARACTERISATION DATA - - - - - - - - - - - - - - - - - - - - - - - 31
3.3.3 - Hydraulic parameters - - - - - - - - - - - - - - - - - - - - - 18 5.4.1 - Basics and general remarks - - - - - - - - - - - - - - - - - 31
5.4.2 - Example of data presentation in tabular form- - - - 31
4 - CHARACTERISTICS OF ROCK MASS - - - - - - - - - - - - - - 19
5.4.3 - Synoptic presentation of rock mass characterisation
4.1 - IDENTIFICATION PARAMETERS- - - - - - - - - - - - - - - - - 19
data and cross-referencing to geological profile- - 31
4.1.1 - RQD- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 19

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 3

 Caracterisation of rock masses useful for the design and the construction of underground structures

1 - INTRODUCTION In both cases, it is vital to arrive at the and identifying singularities and indetermi-
most methodical and comprehensive cha- nate points.
1.1. – PURPOSE OF ROCK racterisation of the rock mass as possible. The geological model is the indispensable
MASS CHARACTERISATION basis for proceeding with the characterisa-
1.2 – GEOLOGICAL MODEL tion of the rock behaviour parameters.
The most important goal in the characte- With this model, the rock can be subdivi-
risation of rock masses is to provide the ded into homogeneous sub-units whose
engineer with qualitative and quantitative 1.2.1 – Geological survey mechanical and hydraulic properties can
data to describe their structure and Before embarking on the stage of a rock subsequently be determined at project
assess their mechanical and hydraulic mass characterisation properly so-called, scale.
properties at a scale commensurate with as defined in these Recommendations, the What governs the size of a sub-unit within
the volume of rock affected by the struc- design process normally begins with a the rock mass is its uniformity of its geo-
tures. The overlying materials (sand, geological survey to situate the project technical properties, producing uniformity
scree, moraine, etc.) are ignored in these area within the general geological setting. of response to the structure to be built. A
Recommendations.
The survey rests essentially on field work by sub-unit may thus occupy part of a geolo-
It is essential to have precise knowledge of engineering geologists using the full gical stage, the whole stage or several
this data for project design, selection of armoury of tools and methods available to stage. It may be homogeneous in terms of
construction methods, support and lining the them: its lithology, jointing, rock stresses, etc.
thickness. The cost of the works is directly
• Desk study to collect published material, Of course, even small features that are only
dependent on these points.
maps and data local singularities in the wider rock mass
Whereas the mechanical properties of the must be treated as individual sub-units and
rock matrix can be determined from labo- • General mapping of the project area, as such, be the subject of geotechnical
ratory tests on small specimens, those of a detailed mapping of outcrops and indica- characterisation.
rock mass measuring several thousand tors, collection of hydrogeological data
cubic metres in size which may contain Subsequently, as the design studies pro-
• Photography at various scales (satellite ceed, the latest results from the geological
within itself many discontinuities and hete-
imagery and aerial photographs) survey will be incorporated from time to
rogeneities cannot be determined directly.
• Geophysical methods: high resolution time into the model.
In situ field tests, whose number is inevita-
bly limited by their high cost, come seismic reflection and refraction, resistivity
and electromagnetic tests, thermogra- 1.3 – GEOTECHNICAL
somewhere between laboratory tests and
phics, ground penetrating radar, etc. CHARACTERISATION OF
the full size structure in terms of scale. They
are instructive but still imperfect for fully SUB-UNITS
• Exploratory boreholes, shafts and adits
ascertaining the mechanical properties of
• Information available from any nearby Characterisation of a homogeneous sub-
the rock mass at the relevant scale.
structures, etc. unit in a rock2 mass always involves deter-
Not being amenable to direct measure- mining the parameters of the rock matrix
ment, the mechanical and hydraulic pro- The geological survey locates the major
and of the discontinuities; with discontinui-
perties of the rock mass must necessarily geological units and their relations, the main
ties, the geological survey must make it
be approached by indirect methods: discontinuities, the tectonic history, etc.
possible to choose the most relevant scale
• either by trying to construct a model of at which they must be analysed and cha-
1.2.2 – Geological model
the rock mass relevant to the size of the racterised with reference to the scale of the
structure under consideration, using test The first step in the characterisation of rock project under design.
data obtained at a smaller scale and the masses requires constructing a tentative
Some homogeneous sub-units as defined
characteristics of the discontinuities, geological model showing the geological
in para. 1.2.2. may consist of more or less
structure of the rock mass complete with its
• or by resorting to current rock classifica- regularly alternating rock layers, each with
constituent units, boundaries, major fea-
tion systems and the mechanical characte- highly contrasting geotechnical properties
tures, heterogeneities and uncertainties.
ristics which can be derived empirically (for example, marl limestone, flysch, etc.)
from them, based on the back-analysis of It will ideally be a three-dimensional which must be analysed separately before
full scale structures, as operated by various conceptual model yielding cross sections proceeding with the characterisation of the
authors. to be used for understanding the structures whole sub-unit.

1
It must be noted that some homogeneous sub-units thus defined are liable to include random heterogeneities (karst cavities for example) unidentifiable
by the exploratory works. It is the engineer's job to decide what steps are to be taken in respect of this risk, after assessing its probability.
2
Rock must be understood in a very general sense. It may be a mass of soft rock or rock crushed by tectonic action to the point where it becomes like a
soil, as well as a highly competent rock mass.

4 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

Characteristic parameters for the rock 2 – MATRIX CHARACTERISTICS exploratory works (drilling) and even more
matrix and discontinuities appear at the so, the cost of building the structure. It is
top of Table 1. Those for the rock mass, PRELIMINARY REMARK. Most laboratory always advisable to perform enough tes-
some of which derive from the former, are tests used to characterise rock matrix para- ting in order to obtain data that can be
listed in the bottom half of the Table. meters are inexpensive compared to field manipulated by statistical methods and

2 CHARACTERISTICS OF ROCK MATRIX 3 CHARACTERISTICS OF DISCONTINUITIES

2.1 IDENTIFICATION PARAMETERS 3.1 IDENTIFICATION PARAMETERS
2.1.1 Common names 3.1.1 Types and origins of discontinuities
2.1.2 Petrography & mineralogy 3.1.2 Description of discontinuities: strike, spacing, persistence,
2.1.3 Alteration of minerals roughness, weathering, width, infill, water
2.1.4 Densities 3.2 CHARACTERISATION OF JOINT SYSTEMS
2.1.5 Volume weights 3.2.1 Directional joint set patterns
2.1.6 Moisture content 3.2.2 Statistical analysis of geometrical parameters for each set:
2.1.7 Porosity orientation, spacing, persistence
2.1.8 Degree of saturation 3.2.3 Lumped jointing density indexes: RQD, ID, FD
2.1.9 Permeability
2.1.10 Ultrasound wave velocities, continuity index

2.2 MECHANICAL PARAMETERS 3.3 MECHANICAL PARAMETERS

2.2.1 Deformability: instantaneous behaviour 3.3.1 Deformation parameters: normal stiffness, tangential stiff-
- Young's modulus - Poisson's ratio ness
2.2.2 Deformability: time-dependent (creep) behaviour 3.3.2 Shear strength parameters: peak strength, residual strength,
2.2.3 Time-dependent (swelling) behaviour dilatancy
2.2.4 Strength 3.3.3 Hydraulic parameters
- Uniaxial compressive strength σ
- Tensile strength σ
- Brittleness index FR
- Point compressive strength (Franklin test)
2.2.5 Triaxial test and failure criteria: Mohr-Coulomb criterion,
Hoek & Brown criterion
2.2.6 Parameters for resistance to excavation: hardness,
drillability, abrasiveness, DRI
2.2.7 Other tests: fragmentability, degradability, LA & MDE tests

4 CHARACTERISTICS OF ROCK MASS

4.1.1 RQD
4.1 IDENTIFICATION PARAMETERS 4.1.2 Alteration and weathering degree
4.1.3 Rock mass continuity index ICM
4.2.1 Rock mass deformability – rock mass deformation modulus
4.2 MECHANICAL PARAMETERS EMas
4.2.2 Limit strength of rock mass
4.3.1 Identification of aquifers
4.3.2 Measurement of initial piezometry
4.3 HYDROGEOLOGICAL CONDITIONS 4.3.3 Measurement of rock mass permeability KM
4.3.4 Gas
4.3.5 Other parameters
4.4.1 Initial state of stress and approximations
4.4 INITIAL STATE OF STRESS IN ROCK MASS
4.4.2 Characterisation of stress tensor
4.4.3 Commentary on field test methods
4.4.4 Classification of stress states
4.5 TEMPERATURE 4.5.1 Geothermal parameters
4.5.2 Temperature assessment methods
Table 1 – Characteristic parameters of rock matrix, discontinuities and rock mass
(Numbers refer to paragraphs in these Guidelines)

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 5

 Caracterisation of rock masses useful for the design and the construction of underground structures

reveal the homogeneity or scatter in the the lithology of the rock concerned. to the walls of major discontinuities in
measurs. Meaningful laboratory test results Reference should be made to the full des- which fluids from deep-lying sources circu-
are essential to draw full benefit from the cription in the description accompanying late. It frequently causes major mineralogi-
information obtained by drilling. the map. cal changes (appearance of special mine-
rals such as chlorite, etc.), usually
It is preferable to use the terms in
2.1 – IDENTIFICATION Appendix 1 and avoid employing unusual
accompanied by significant changes in
PARAMETERS mechanical properties.
complex names.
• Weathering working down from the sur-
2.1.1 – Common names 2.1.2 – Petrography and face to sometimes considerable depth.
mineralogy Processes include solution of gypsum, cal-
Rock names are based on chemical and cite, etc., mechanical disruption (increased
mineralogical composition, texture and the A petrographic description covers the fol-
microcracking) and mineralogical changes
way their were formed. There are three lowing observations with the naked eye or
producing clay minerals.
main classes of rock: igneous, metamor- magnifying glass or (preferably) by inspec-
phic and sedimentary rock (Appendix 1). tion of thin sections under the microscope: The intensity of alteration and weathering
of the matrix can be quantified by minera-
Igneous rock is solidified magma. • Identification of minerals present
logical analysis and indirectly by tests such
Solidification at depth produces plutonic • Size and arrangement of the minerals as the methylene blue test and measure-
rock which solidified slowly and permitted (texture) ment of ultrasound wave velocity.
crystals to grow large enough to be seen • Proportions of the different constituents A clear distinction must be made between
with the naked eye; the most common
• Voids and discontinuities (pores and fis- the degree of alteration of rock as taken
example is granite. Extrusive rock is for-
sures). from a borehole or at the tunnel face and
med from magmas emerging directly at
its potential of alterability when exposed to
the Earth's surface; few crystals can be Mineralogical analysis of the constituents the atmosphere.
seen by eye because of the rapid cooling establishes the mineral composition of the
of the material. The most widespread rock and yields information on its proper-
extrusive rock is basalt. ties such as weathering potential, swelling
Sedimentary rock forms at the surface, on potential, ability to stick, abrasion poten-
land or under water, by deposition of origi- tial, etc.
nally near-horizontal beds. Sedimentary The mineralogical analysis is usually perfor-
rock subdivides into: med by X ray diffraction on a small powde-
• Detrital rocks, resulting from the deposi- red specimen. It allows the identification of
tion of debris from pre-existing rocks resul- the minerals present and, after interpreta-
ting from erosion and transport processes tion, yields the semi-quantified composi-
(running water, glaciers, wind); the most tion. Special preparation is needed if it is
widespread representatives are sandstone suspected that swelling clay minerals might
and the argillaceous rocks. be present. Photography 1– Hydrothermal alteration – Granite
(Ghangzou China)
• Physical/chemical and/or biogenic rocks Additional quantitative determinations of
formed by precipitation of ions in solution CaCO3, silica, sulphates, organic matter,
and/or living matter; the commonest are etc. refine the identification.
2.1.4 – Densities (French standard
carbonate rocks and saline rocks, still called The clay fraction, if present, must be cha- P 94-410-1/2/3)
evaporites. racterised from the Atterberg limits. A Different densities (M.L.-3 dimensions) are
Metamorphic rocks are the result of pro- methylene blue absorption test will esti- applicable according to the condition of
found transformation in the solid state of mate the activity of the clay fraction (French the material.
pre-existing sedimentary or igneous rocks standard NF P 94-068).
• Natural density ρ = m/v
by elevated temperatures and/or pres-
sures. They often exhibit schistosity or 2.1.3 – Alteration of the minerals Ratio between dry mass md of the oven-
foliation accompanied by lineation. The in the rock matrix dried sample and the volume V of the
commonest are schist and gneiss in which Alteration of the matrix is the result of phy- sample including any air it contains.
the minerals are strongly oriented. Marble sical/chemical changes in the constituent • Dry density ρd = md /v
and quartzite are massive, completely rock minerals. It is usually associated with Ratio between dry mass md of the oven-
recrystallised rocks in which the orientation major changes in the physical and mecha- dried sample and the volume V of the
of the minerals (calcite or quartz) is hardly nical properties of the rock. Some minerals sample including any air it contains.
visible to the naked eye. are subject to dissolution (e.g. calcite, gyp-
sum), other to decompositione (e.g. bio- • Density of solid particles ρs = ms/vs
It is important to bear in mind possible
variations in the facies of rocks belonging tite, plagioclase). As a general rule, the Ratio between the dry mass of solid par-
to the same geological stage and the fact rock loses cohesion. The process is usually ticles ms in a powdered specimen and the
that some common names deriving from subdivided into volume vs occupied by the particles (mea-
the geological map do not always tally with • Hydrothermal alteration, usually confined sured in a pycnometer). This characteristic

6 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

of the solid phase of the rock material is Sr = (vw/vv) x 100 velocity Vp* from the table in Appendix 3
directly dependent on the mineral compo- Rock is described as 'dry' when Sr = 0. It is showing maximum theoretical wave veloci-
sition of the rock. Appendix 1 lists values ties for some rocks, assumed to be sound
'saturated' when Sr = 100%.
for the more common minerals. without pores and fissures.
2.1.9 – Permeability The continuity index for the same petro-
2.1.5 – Volume weights
The permeability k of a rock sample is des- graphic composition falls as pore porosity
Corresponding unit weights (M.L -2.T 2 increases; the trend is even more marked
cribed by a coefficient relating the flow Q
dimensions) γ, γd and γs to he above unit as microcrack porosity increases.
passing across a surface S under a hydrau-
masses are obtained by multiplying the
lic head gradient i (Darcy's law). Continuity classes based on IC value are lis-
unit masses by the acceleration due to gra-
Q/S = k x i ted in Table 3.
vity g = 9.81 m/s2:
γ=ρxg The dimension of k is a velocity (L.T1).
CLASS CONTINUITY DESCRIPTION
The permeability of the rock matrix is stron- INDEX IC
2.1.6 – Moisture content (French gly influenced by microcracking (intercon-
standard P 94-410-1) IC 1 IC > 90 % Very high continuity
nected voids) and therefore varies with the
The moisture content by weight w is the state of stress. The proper choice of repre- IC 2 75 %< IC < 90 % High continuity
ratio, expressed as percentage of the sentative samples and their pre-test condi- IC 3 50 %< IC < 75 % Moderate continuity
mass of water mw to the mass of the dry tion is particularly important. Laboratory
material md: tests are done with special longitudinal, IC 4 25 %< IC < 50 % Low continuity
w(%) = (mw/md) x 100 radial, etc. permeameters or triaxial appa- IC 5 IC < 25 % Very low continuity
ratus. If permeability is found to be aniso- Table 3 – Rock matrix continuity classes
2.1.7. Porosity tropic, tests should be made in several
(French standard P 94-410-3) directions.
The porosity n is the ratio, expressed as Knowledge of the matrix permeability is 2.2 – MECHANICAL
percentage of the volume of voids vv to the essential only for some underground pro- PARAMETERS
total volume of the rock sample v: jects (mined storage, waste disposal by
containment, etc.). The mechanical parameters relevant to
n (%) = (vv/v) x 100 geotechnical classification and to the
The porosity is governed mainly by the 2.1.9 – Ultrasound wave velocity choice and optimisation of tunnelling tech-
presence of globular voids (pores) and (French standard p 94-411) – niques and plant are determined in the
much less by fissures (flat, thin voids). Some Continuity index laboratory from representative samples.
voids may be inaccessible to saturation Ultrasound wave velocity yields informa- Attention must be given to any anisotropy
water (occluded voids). tion on alteration and weathering and/or in measured properties. Parameters descri-
Values for porosity classes are listed in fissuration and porosity. bing mechanical behaviour very often dif-
Table 2. fer according to the orientation of the
Measuring wave velocity in several diffe- sample with respect to bedding (in sedi-
rent directions may reveal anisotropy due mentary rocks) or foliation (in metamorphic
to preferential orientation of microcracks or rocks, see example in figure 1). With some
POROSITY
CLASS N VALUES DESCRIPTION rock structure. rocks, the anisotropy ratio, defined as the
P1 0%<n<1% Very low porosity There are two types of wave: ratio between maximum and minimum
• Compression ( or longitudinal or P) waves values of a parameter, measured at diffe-
P2 1%<n<5% low porosity
noted Vp rent orientations, may be in excess of 5.
P3 5 % < n < 15 % Moderate porosity
• Shear (or transverse or S) waves, not rou-
P4 15 % < n < 30 % High moderate
porosity tinely measured, noted Vs.
P5 n > 30 % Very high porosity The continuity index IC of rock is defined
as the ratio between Vp measured on
Table 2 – Porosity classes for rock matrix
samples and the theoretical Vp* derived
In the majority of sedimentary rocks, a dry from the mineral composition of the
unit mass ρd of the order of 2.7 t/m3 is a sample:
good indicator of the porosity. IC (%) = 100 x (Vp/Vp*)

2.1.8 – Degree of saturation Vp* is the harmonic mean of the Vpi wave
velocities in the constituent rock minerals
The degree of saturation with water Sr is (Appendix 2) multiplied by their proportion
the ratio, expressed as percentage of the by volume ci:
volume of water in the sample vw to the
volume of voids vv; it is the percentage of 1/Vp* = ∑ ci/Vpi
the pore space filled with water. An approximation is usually possible. It Figure 1 – Example of anisotropic uniaxial compressive
consists of estimating the theoretical wave strength (Lias shales, Grand'Maison, France, 1980)

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 7

 Caracterisation of rock masses useful for the design and the construction of underground structures

Mechanical tests should be made in seve- ceous rocks (marl, claystone) and some car- The creep test may be performed under
ral directions; this must always be done bonate rocks such as chalk. uniaxial or triaxial pressure and drained or
with metamorphic rocks. 2. They may be the consequence of undrained conditions, preferably at several
damage. In any rock, time-dependent different deviator stresses, to determine
2.2.1 – Deformability: effects more or less appear when the the following parameters (figure 2):
instantaneous behaviour
microcracking threshold is exceeded.
2.2.1.1 – Young's modulus 3. They may be due to changes over time
(French standard P 94-425) in the pressure of fluids in pores and fis-
In a uniaxial compression test, the Young's sures, due to changes in boundary condi-
modulus of elasticity E is defined as the tions of flow patterns caused by construc-
slope of a loading/unloading cycle on the tion of the works (effect of consolidation
axial stress/strain curve σ1 = f(ε1) at half the and fluid flow pressures).
failure stress of the sample σc. Only time-dependent effects associated
Classes based on stiffness values (the with creep mechanisms 1 and 2 are consi-
reciprocal of deformability) are listed in dered here. They produce viscoelastic
Table 4. (reversible) or viscoplastic (irreversible) res-
ponses in the rock matrix. They have
nothing in common with responses due to
CLASSES YOUNG'S DESCRIPTION fluid flow (mechanism 3).
MODULUS Figure 2 – IIncremental creep test
Time-dependent behaviour is more mar-
DE 1 E > 50 GPa Extremely stiff matrix ked as higher temperatures reduce visco- I No creep: applied stress lower than creep threshold
II Transient (primary) creep decreasing over time:
DE 2 20 GPa < E < 50 GPa Very stiff matrix sity. This is particularly noticeable in evapo- applied stress slightly greater than creep threshold
rites. High stresses have the same effect, III Creep slowing down over time: applied stress lower
DE 3 5 GPa < E < 20 GPa Stiff matrix especially deviator stresses. than long-term strength
IV Three-phase primary, secondary (stationary) and
DE 4 1 GPa < E < 5 GPa Fairly stiff matrix The long-term performance of under- tertiary (culminating in failure) creep: applied stress
DE 5 0,1 GPa < E < 1 GPa Low stiff matrix ground structures may be affected by greater that long-term strength
changing stresses and strains, and progres-
DE 6 E < 0,1 GPa Very low stiff matrix sive loss of strength. The magnitude of
these processes is frequently different in • Creep threshold, the state of stress below
Table 4 – Rock matrix stiffness classes
(reciprocal of deformability) situ from in the laboratory, although labora- which creep is negligible (the threshold
tory tests do yield a significant approxima- may be near-zero for some evaporites).
tion of the importance, pattern and order • Creep rate, which generally increases
2.2.1.2 – Poisson's ratio of magnitude of time-dependent beha- with deviator stress.
(French standard P 94-425) viour to be expected in situ.
• Creep acceleration threshold beyond
In the uniaxial compression test, the
2.2.2.2 – Laboratory characterisation of which "tertiary creep" occurs and culmi-
Poisson's ratio n is defined as the ratio bet-
time-dependent behaviour: creep test nates in failure; it characterises the long
ween the slopes of the straight-line limbs
term strength of the rock. In rocks exhibi-
of the σ1 = f(ε3) and σ1 = f(ε1) curves. In studying time-dependent rock beha-
viour in the laboratory, the commonest test ting no (or little) creep under normal civil
ν = dε3/ dε1 engineering conditions (granites, etc.),
is the creep test which is a stress-controlled
ε1 = axial strain test to measure strain under constant long term strength is of the order of 90% of
the strength measured in quick tests (such
ε3 = transverse strain. stress. The relaxation test, a strain-control-
led test which measures stress changes as the standard uniaxial compressive
Values of the Poisson's ratio for different strength test); in chalks, it is of the order of
under constant strain, is not often used.
rocks generally fall between 0.15 and 0.40. 50%; it falls to 30% or less for evaporites.
The creep test is usually preferred. It offers a
Creep rates cover a very extended range,
2.2.2 – Deformability: reasonable simulation of conditions obtai-
depending on the stress ranges concerned:
time-dependent behaviour related ning around underground openings; after
to creep the transient tunnelling period, the state of • The fastest rates (>10-5s-1) cover the range
stress around the opening thereafter of instantaneous strain as measured in the
2.2.2.1 – Definition remains substantially unchanged over time. laboratory.
Time-dependent effects take the form of If laboratory tests characterising time- • The slowest rates (<10-11s-1) are scarcely
changes in strain and/or stress over time. dependent or long-term rock behaviour measurable but represent deformation
They have three causes: are to have maximum significance, test over geological time.
1. They may be related to an intrinsic rheo- conditions (temperature, stress range, Creep rates relevant to actual tunnels,
logical property of the rock whose defor- degree of saturation, drained or undrained chambers, storage cavities, etc. will lie
mation under constant load increase over boundary condition, etc.) must reflect somewhere between the two extremes.
time. It is mainly encountered in evaporites actual in situ conditions of the structure Feedback from real projects is still scarce
(rock salt, potash, gypsum, etc.), argilla- and the host materials.

8 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

and possible rates cover such a wide range gical analyses for the proportion of swel- In practical terms, the first step should be
(being governed by the characteristics of ling clay mineral and/or anhydrite and gyp- to measure the axial swelling pressure at
the rock mass and the underground struc- sum present. constant volume. Next, axial pressure is
ture) as to make any forecast of time- • If anhydrite is present, microcracking measured versus axial strain (Huder-
dependent creep around a real tunnel should be investigated, since water take- Amberg type), the specimen being wetted
highly problematical (see para. 4.2.1.5 for the first time with axial pressure sub-
up is directly linked to the exchange sur-
below). stantially equivalent to the previously
face area available.
determined swelling pressure.
2.2.3 – Time-dependent behaviour Additional information can be obtained
related to swelling with the electronic microscope to gain a Axial swelling pressures may vary greatly in
better insight into the distribution of the different materials, from near-negligible
2.2.3.1 – Swelling potential clay flakes and any other special features (less than 0.1 MPa) to several MPa in some
liable to favour or oppose swelling (such as marls. There is frequently severe scatter in
Rock swelling is due to the rock increasing
smectite being encapsulated in a calcite the swelling pressures measured on any
in volume over time, simultaneously with
increasing moisture content (wetting under envelope for example). given material. This is why it is advisable to
saturated or unsaturated conditions). perform several tests, regardless of the
Changes in the state of stress is a promo- 2.2.3.3 – Quantification of swelling type of test used to quantify swelling
ting factor. When rock expansion is potential potential.
constrained, high stresses may result. To confirm the risk of swelling, mineralogi-
2.2.3.4 – Comments
Rock swelling has two causes: cal analysis must be supplemented with
laboratory tests. Materials containing clay minerals (clay,
1. Water may be taken up by hydrophilic marl, molasses) frequently exhibit conside-
minerals, mainly swelling clay minerals such The International Society for Rock
rable anisotropy, due to the way they were
as smectite , some hydroxides and sul- Mechanics ISRM recommends three labo-
formed. This anisotropy may lead to very
phates. ratory tests to characterise rock swelling
different swelling potentials when tested in
potential (Int. J. of Rock Mech. and Min.
2. Change from anhydrite (CaSO4) to gyp- different directions, especially the direc-
Sci., 1999, vol. 36, 291-306):
sum (CaSO4,2H2O). tions parallel and perpendicular to the
1. Measurement of axial swelling pressure plane in which the clay particles were
Smectite-type minerals are relatively com-
at constant volume (determination of swel- deposited. Tests must be made in these
monplace and might be present whenever
ling pressure σg). two directions for a proper characterisation
the rock contains clay minerals: clay, marl,
molasse, fault gouge, karst infill, weathered 2. Determination of unrestrained axial and of swelling.
igneous and metamorphic rock. radial swelling strain. The composition of the water used in the
Anhydrite and gypsum occur either as 3. Measurement of axial pressure versus test may exert a strong influence on the
large masses in sedimentary beds or intru- axial strain (to determine swelling index Cg). development of swelling. Certain chemi-
ded into large tectonic discontinuities, or cals may promote or inhibit swelling. The
The third test derives directly from the test water should therefore be clearly iden-
again, finely scattered within marl or other
Huder-Amberg test developed in the early tified. Swelling may develop very diffe-
material.
seventies to characterise swelling in soils rently in situ, depending on whether the
Three conditions must be fulfilled for swel- (cf. Appendix 4). water comes from the surrounding rock or
ling to occur:
The swelling test for soils described in from some external source (from inside the
1. Potentially swelling minerals in the rock. French standard P 94-091 uses the oedo- tunnel).
2. Available water. meter to measure swelling on four speci-
mens assumed to be identical. The speci- 2.2.4 – Mechanical strength
3. A state of stress that permits volume
mens are wetted and subjected to four
increase. 2.2.4.1 – – Unconfined compressive
different levels of axial stress. The test
consists of measuring the thickness strength σc or uniaxial compression test
2.2.3.2 – Identification of swelling (French standard P 94-420
potential increase ∆h/h of each of the four speci-
mens and correlating thickness change The failure stress in uniaxial compression is
IIt is strongly recommended to identify defined as
with the corresponding axial stress levels
the risk of swelling at a very early stage, as
by fitting a straight line on a semi-log plot σc = Fmax/A
follows:
(∆h/h – log σ).
• Qualitative observation of behaviour by Fmax = maximum axial force reached in the
immersing specimens in water and seeing The Huder-Amberg test uses the same test
how fast they break up. type of interpretation but has the advan-
tage of requiring only one test specimen, A = area of the pre-test circular cross sec-
• Methylene blue tests to determine the and this limits the ever-present problem of tion of the test specimen.
specific surface and clay character of the scatter inherent in irregularities in material Rock strength classes used in the present
material. type and condition, when using multiple Recommendations3 and ISRM appear in
• Total or semi-quantitative (R.X.) mineralo- specimens. Table 5.

3
They differ from the classes used in the AFTES 1978 Recommendations

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 9

 Caracterisation of rock masses useful for the design and the construction of underground structures

may range from 25cm to 100cm. Test spe-

cimens are usually pieces of rock taken
from 50mm dia. cores, although charts are
available to correct results from other core
sizes (Broch & Franklin, Int. J. of Rock
Mech. and Min. Sc., 1972, vol 9, 669-697).
Test results are expressed as a strength
index Is (in MPa):
Is = F/D2
F = load at failure
D = specimen diameter or distance bet-
ween points.
The index obtained with a 50mm diameter
is written Is50.
Photo 3 – Specimen after failure in triaxial test The point load compression test can be
(French standard p 94-422) performed with very lightweight apparatus
on site. Correlations yield an estimate of
The test breaks a cylinder of diameter D rock unconfined compressive strength:
and height H by applying a compressive 20 Is50 < σc < 27 Is50
Photo 2 - Uniaxial compression test load F at diametrically opposed sides of
the cylinder. The stress at failure of the However, the point load compression test
Brazilian test specimen σtb is given by the can never be considered as a substitute for
UNIAXIAL equation the uniaxial compression test.
COMPRESSIVE CLASS DESCRIPTION
STRENGTH σc
σtb = 2 Fmax/π x D x H 2.2.5 – Triaxial test and
Fmax = load at failure failure criteria
σc > 200 MPa RC1 Extremely strong matrix
100 MPa < σc < 200 MPa RC2 Very strong matrix H = cylinder height 2.2.5.1 – Triaxial test
50 MPa < σc < 100 MPa RC3 Strong matrix D = cylinder diameter. (French standard P 94-423)

25 MPa < σc < 50 MPa RC4 Moderately strong The triaxial test is an axial compression test
2.2.4.2 – Brittleness index FR
in which a constant confining stress σ3 is
5 MPa < σc < 25 MPa RC5 Low strong matrix The brittleness index FR is defined as the applied to the specimen. The axial com-
1 MPa < σc < 5 MPa RC6 Very low strong matrix ratio σc/σtb. pressive stress is continued to failure.
σc< 1 MPA RC7 Extremely low strong It is useful as characterising the drillability Determination of the failure criterion for a
matrix and failure type for hard rocks (σc > 25 MPa). rock involves performing several triaxial tests
Table 5 – Uniaxial compressive strength classes FR usually ranges from 5 to 30. Brittleness with increasing confining stresses. There
classes are listed in Table 6. must be at least four test runs, including a
uniaxial compression test in which σ3 = 0.
Soft rock and stiff soils often classify as
Class RC6 and RC7. The full σ1 - σ3 = f(ε1) curve is plotted for
BRITTLENESS
CLASS
INDEX FR VALUES
DESCRIPTION each test at a different confining stress to
Rock tensile strength σtb is determined by find the peak value of the deviator failure
the indirect Brazilian test (figure 3) using FR 1 FR >25 Very brittle matrix stress σ1 - σ3, plus, possibly, the residual
the procedure described in French stan- FR 2 15 < FR < 25 Brittle matrix (constant) stress at which the specimen
dard P 94-422. shears along the failure plane.
FR 3 10 < FR < 15 Moderately brittle matrix
FR 4 FR < 10 Low brittle matrix Measurement of axial and transverse strain
in the course of the test leads to values for
Table 6 – Rock matrix brittleness classes
the Young's modulus and Poisson's ratio
under confining stress conditions.
2.2.4.3 – Point load compression test, This is the standard soils mechanics crite-
so called– Franklin test rion written as
(French standard P 94-429 τ = c + σn x tan ϕ
The point load compression tests involves • σn is the normal stress and and τ is the
breaking rock fragments of indeterminate shear stress on the failure surface
shape or core pieces between two cones
with spherical tips (dimensions standardi- • c is cohesion
Figure 3 – Stress state at failure at specimen centre sed). Specimen thickness between points • ϕ is the internal friction angle.
in compression test

10 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

In terms of the principal stresses, with σ1 > σ3, parameter mi is very close to the brittleness Rock hardness classes based on the CER-
the Mohr-Coulomb criterion is written index FR: CHAR-INERIS rules are listed in Table 7.
σ1 = [σ3(1 + sin φ) + 2c(cos φ)] / (1 – sin φ) mi ≈ σci /σti = FR
2.2.6.2 – Abrasiveness
This limit strength criterion can apply to the
elastic limits, peaks or plateaus on the stress 2.2.6 – Parameters for resistance
HARDNESS DESCRIPTION
to excavation CLASS
VALUES
strain curves from triaxial tests (total or effec-
tive stresses, drained or undrained tests). The parameters examined in this section DU 1 > 120 Extremely hard matrix
concern the response of rock to various
The Mohr-Coulomb criterion may suit the DU 2 80 - 120 Very hard matrix
means of breaking it up (its decohesion).
mechanical behaviour of certain rocks in
They are useful for assessing rock in res- DU 3 40 - 80 Hard matrix
the moderate confining stress range. More
pect of excavating and crushing tech-
generally, it may be acceptable for repre- DU 4 20 - 40 Moderately hard matrix
niques, in order to refine forecasts as to
senting the behaviour of a given rock DU 5 5 - 20 Soft matrix
within a certain limited range of confining • mechanised excavation, through hard-
stresses (linearisation of a parabolic crite- ness and drillability tests, DU 6 <5 Very soft matrix
rion). • excavating tool wear and consumption, Table 7 – Rock matrix hardness classes based on
through abrasion tests, CERCHAR-INERIS test results
2.2.5.2 – Hoek & Brown criterion
• crushing performance, through fragmen-
This parabolic criterion is well suited to the tation tests.
mechanical behaviour of rock and is usually The abrasion potential of a rock is determi-
used for these materials. Test results are interpreted with the aid of ned by its mineralogical composition,
multi-criteria correlation models which also especially the percentage of quartz it
The criterion for intact rock is written refer to other parameters (mineralogy, contains, and its intergranular cohesion
σ1 = σ3 + σci x [(mi σci) + 1]1/2 mechanical properties, discontinuities), and grain size.
in which investigated in another location.
Abrasiveness can be characterised by two
• σci is the uniaxial compressive strength of 2.2.6.1 – Hardness and “ drillability” conventional, standardised indexes; it must
the intact rock material be recognised that they bear no relation-
These tests can be classified under three ship to each other.
• mi is a constant dependent on rock type. headings based on the techniques used.
From the standard expression of a parabo- 1. Tests assessing the penetration of drill 2.2.6.2.1 – Point scratch test – French stan-
lic criterion expressed in terms of uniaxial bits and cutter tools, such as the CERCHAR, dard P 94-430-1
compressive σ ci and σ ti tensile failure Siever and other tests. The first index, AIN4 comes from the CER-
stresses (Appendix 5), it can be seen that CHAR-INERIS point scratch test. Test
2. Static indentation tests
which assess the mark left by a results are expressed as an abrasiveness
point applied to the rock, such index characterising the ability of a rock to
as the punch test, Vickers, cause wear of the cutting tool.
Knoops and Schreiner tests, Rock abrasiveness classes derived from this
micro-indentation test, etc. index are listed in Table 8.
3. Rebound tests, in which the ABRASIVENESS
DESCRIPTION
rebound of a known mass hit- CLASS INDEX AIN
VALUES
ting the rock is measured
(Schmidt hammer). AIN 1 >4.0 * Extremely abrasive matrix
The commonest test is the AIN 2 2.0 – 4.0 Very abrasive matrix
penetration test derived from AIN 3 1.0 – 2.0 Abrasive matrix
the test first described by
French research institute CER- AIN 4 0.5 – 1.0 Low abrasiveness matrix
CHAR: "Drill bit penetration AIN 5 < 0.5 Very low abrasive matrix
resistance index" – French
standard P 94-412. *Quartz and gemstones register higher than 6

It characterises rock hardness Table 8 – Rock matrix abrasiveness classes based on

results of CERCHAR-INERIS test
via its resistance to penetration
by a drill bit under standard
test conditions. The hardness 2.2.6.2.2 – Rotating drilling bit test (French
standard P 94-430-2)
test is suited to fine grained
rocks of moderate to low The other index is derived from the LCPC
strength. abrasiveness test. This test is suitable for
rocks whose tensile strength is greater than
Photo 4 – Triaxial test apparatus

4
CAI (CERCHAR Abrasiveness Index) in English-speaking countries

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 11

 Caracterisation of rock masses useful for the design and the construction of underground structures

1 MPa. The result is expressed by the abra- 3 – CHARACTERISTICS OF

siveness index ABR calculated from wear of DISCONTINUITIES
a drill bit rotated in a body of 4/6.3 aggre-
gate obtained from the rock under test. The system of discontinuities in the rock
Rock abrasiveness classes based on the mass must be investigated in detail at pro-
LCPC test are listed in Table 9. totype scale of the project, mainly through
statistical analysis in order to consider the
ABRASIVENESS natural variability in their geometrical and
CLASS DESCRIPTION mechanical parameters. Different data
INDEX ABR
VALUES acquisition methods are appropriate to dif-
ABR 1 ABR > 2000 Very highly abrasive ferent scales of observation:
ABR 2 1500 < ABR < 2000 Highly abrasive • aerial photography and surface geophy-
ABR 3 1000 < ABR < 1500 Moderately abrasive
sics at regional scale
ABR 4 500 < ABR < 1000 Low abrasiveness • outcrop and borehole surveys at local scale.
ABR 5 0 < ABR < 500 Very low abrasiveness When core drilling, it is always recommen-
ded to accompany the geologist's logs
Table 9 – Rock matrix abrasiveness test based (geological drill logs which include records
on results of LCPC test
from the various borehole loggings and
Photo 5 – Rock mass with complex tectonic history.
joint mapping) with colour photographs of Tibi region, Spain
2.2.6.3 – Norwegian tests: DRI index all the core boxes; photographs should
The Norwegian Drilling Rate Index defined include a scale, colour chart and legible
by the University of Trondheim (Movinkel & identification markings. Photographic • Joints properly so called, are discontinui-
Johannessen 1986) combines the results of records may indeed provide the engineer ties between two rock compartments
an S20 fragmentation test and an SJ drill with abundant valuable information, espe- without any apparent relative movement
bit penetration test. The DRI chart and a cially about discontinuities. However, they between them. The joint may be tight or
description of the S20 and SJ tests appear could not replace the direct inspection of open with generally planar, relatively
in Appendix 6. the cores which is essential , even if it is smooth walls, and may extend over a dis-
often difficult. tance measured in decimetres to deca-
2.2.7 – Other tests metres. They often occur as joint sets run-
2.2.7.1 – – Fragmentability (French stan- 3.1 – JOINT IDENTIFICATION ning in two or three directions.
dard P 94-066) – Degradability (P 94-067) PARAMETERS • Faults are the result of relative movement
These standard tests characterise the sus- between the two rock compartments bet-
ceptibility of the rock to fragment and 3.1.1 – Types and origins of the ween which they lie, caused by the stress
degrade under the action of drilling and discontinuities field during the relevant tectonic episode
excavating machines. They are mainly used (normal, reverse and strike-slip faults).
The term discontinuity is used in rock
when rock cuttings are to be used as Persistence is very variable (metres to seve-
mechanics in a very general sense, to desi-
above-ground building material, but they ral kilometres long) and they frequently
gnate any interuption of the continuity in
are undoubtedly useful in tunnelling work contain infilling material (gouge) with poor
the rock material with its attendant mecha-
to characterise spoil muck. mechanical properties.
nical, hydraulic and thermal properties.
The surface can be treated, over a certain • Schistosity planes cause the rock to break
2.2.7.2 – Los Angeles test (French stan- distance, as a plane displaying zero or low into thin parallel layers under the action of
dard P 18-573) – Micro-Deval test in tectonic stresses. Persistence is variable
presence of water (P 18-572 tensile strength.
and small unitary schistosity planes may
It will be remembered that the Los Angeles Discontinuities represent a wide variety of
form surfaces persisting for very long dis-
test characterises the resistance of rock to types of surfaces whose geological identifi-
tances through localised failure of rock
fragmentation while the micro-Deval cation conveys important information on
bridges between closely-spaced parallel or
(MDE) test characterises the rock's resis- some of their geometrical and mechanical
stepped discontinuities.
tance to wear in the presence of water. features (parameters). Instead of the
These are essential parameters when dea- conventional terminology of joints, frac- • Fracture zones are assemblages of dis-
ling with hard rocks, for characterising the tures or cracks, preference is given to the continuities of variable persistence and
possibility, when applicable, of using muck following terms: orientation, organised in a more or less pla-
as something more elaborate than plain fill nar pattern.
• Bedding planes are surfaces separating
(ballast, road-making and concrete aggre- sedimentary rock strata. They are very per- • Lithological contact surfaces between
gate, road foundation courses, etc.). sistent and may contain clay material resul- host rock and veins often form virtual dis-
ting in very low shear strength. continuities, although they may be real sur-

12 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

faces where differential weathering has

occurred.
Note : If in French, discontinuity is the generic
term, in English, “joint” is the term currently used
for discontinuity in this recommendation. For
example, joint set means set of discontinuities.

3.1.2 –Description of discontinuities Photo 6

Sangatte grey chalk
Discontinuities within a homogeneous zone
are systematically mapped in order to charac-
terise, statistically, a joint system usually
consisting of several joint sets with different
strikes and possibly a few singular units which
may be important from the rock mechanics
viewpoint (faults, altered veins, etc.).
These observations are made along a
scanline marked on natural outcrops, tun-
nel walls or the sides of excavations, or
from boreholes, whether or not cores have also determines the shapes of the indivi- 4. Joint wall roughness and waviness: mea-
been recovered (down-the-hole viewing dual blocks comprising the rock mass, and sured respectively in millimetres or centi-
techniques). thereby, the anisotropy which will govern metres, and in decimetres to metres. These
hydraulic and mechanical behaviour. are crucial parameters because they
In order to ensure these measures are govern the process of dilatancy and there-
representative, surveys must be plotted in 2. Spacing: is the perpendicular distance
(modal or average) between adjacent dis- fore the joint shear strength. Although diffi-
several spatial directions and must cover a
continuities in a set. In fact, the distance cult to measure, every effort must be made
sufficiently large volume, commensurate
between two successive intersections of to estimate them (cf. para. 3.3.2 Shear
with the average joint density. Lastly, it is
the joint trace with the survey line can be Strength Parameters).
important that the method used to collect
the data be clearly described. measured in the field. This measurement is 5. Joint wall weathering: This is an impor-
frequently biased because it depends on tant parameter especially when the discon-
The full procedure for the systematic analy- the joint persistence (for any given number tinuity walls are in direct rock to rock intact,
sis of joint systems is based on characteri- of traces on a surface, the longer ones have because it governs the deformability and
sing each discontinuity with the following more chance of being cut by the line of the the possibility of dilatancy affecting the
eight parameters: survey and therefore appear to be more shear strength. The degree of alteration
1. Strike: describes the position of the dis- closely spaced) and the direction of the can be assessed directly in the field from
continuity plane in space with respect to survey line. the description of the weathering pro-
North. Two conventions are used (figure 4): 3. Persistence: Joint persistence or size ducts, their thickness and the Schmidt
• The first, based on the direction of the concerns the total area of the discontinuity hammer test (Appendix 9).
horizontal line of the plane α (0-180°), the in all directions. It is an important parameter 6. Joint width or aperture: Size of the gap
dip angle β (0-90°) and the direction of the because, along with joint spacing, it governs between the joint walls measured perpen-
dip vector, is normally used in field work. the connectivity of the discontinuity net- dicularly to the joint plane.
• The second, which uses the azimuth of work and therefore the permeability of the
rock mass and the volume of the intact rock 7. Infill: The nature of the material filling
the dip vector αp (0-360°) and dip angle the discontinuity or coating the walls must
β (0-90°), is the method recommended by blocks. It cannot be measured directly, but
can be quantify by observing the disconti- be characterised, along with its thickness
AFTES for plotting results. and its mechanical properties.
nuity trace lengths (intersection line of the
Strike is the basic parameter for establi- discontinuity and the surface of exposure. It 8. Presence of water: Presence of damp
shing the initial identification of joint sets. It cannot be measured in boreholes. spots or water flow.

3.2 – CHARACTERISATION
OF JOINT SYSTEMS

3.2.1 – Directional joint set

patterns
Discontinuities are not arbitrarily orien-
ted in rock, they frequently occur as
sets, in numbers determined by the
geological and mechanical processes
Figure 4 – Attitude of a disontinuty in space

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 13

 Caracterisation of rock masses useful for the design and the construction of underground structures

occurring at the time of rock formation and ORIENTATION OF DISCONTINUITIES

the tectonic history of the rock body. The DESCRIPTION OF TUNNELLING
CLASS Angle δ between direction of CONDITIONS
distribution of a population of discontinui-
dip vector αp and axis Dip β
ties into directional sets is investigated
of heading
from the azimuth and dip parameters but
other geometric parameters such as aper- OR 1 Indeterminate 0° to 20° With subhorizontal beds
ture and persistence may also be relevant a 0° to 30° 20° to 90° (a) with dip
OR 2 Cross-bed
in identifying joint sets. b (b) against dip
The very concept of joint sets however is a OR 3 30° to 65° 20° to 90° Intermédiate conditions
simplification of reality which may some- (a) moderate dip
a 65° to 90° 20° to 60°
times be excessive or unjustified. This type OR 4 With strike
of analysis must therefore be conducted b 60° to 90° (b) steep dip
with prudence, making use of the geolo- Table 11 – Joint orientation classes with respect to tunnel axis
gist's experience. In all cases, it is neces-
sary to work according to joint type, distin- 3.2.2.2 – Joint spacing in each joint set
guishing for example between bedding
planes, foliation or schistosity, joints proper The histogram of joint spacing ei in a joint
and faults. set can be readily obtained from the dis-
tances di between discontinuities in the
The most frequently used analytical same set intersecting the survey lines, with
approach is the stereographic projection reference to the angle θ between the sur-
(see Appendix 7 for more details). On the vey line and the normal to the mean plane
basis of field observations and stereogra- of the joint set (ei = di.cosθ). The next step
phic plots, the pattern of joint sets can be is to calculate the mean value ES and stan-
described with the help of Table 10. dard deviation of the spacing, and various
modal values if they emerge clearly from
3.2.2 – Statistical analysis the histogram.
of geometrical parameters In bedded rock masses, spacing is equal to
for each joint set bed thickness.
Once the directional joint sets (if any) have Joint spacing classes are listed in Table 12.
been identified, statistical analysis of each
set can proceed, by means of distribution 3.2.2.3 – Joint set persistence
Figure 5 – Joint orientation. Stereogram and polar plot
histograms of geometrical parameters for one joint set Joint persistence must be analysed with
such as strike, persistence and spacing. caution. It can only be estimated from
Sectors in the stereogram correspond to
One can then calculate the means and the geometrical locus of the poles of joint
standard deviations for each parameter, planes oriented according to OR classes
and a distribution function may be fitted. from Table 11. See also sketches in figure 6

3.2.2.1 – Joint set orientation

joint set dictate the tunnelling
The orientation of each joint set must be method, as described in Table 11.
considered in relation to the direction of The stereogram in figure 5 is a gra-
tunnel driving. The dip angle β and the phical representation of these
angle δ between the axis of heading A and classes, with explicit examples in
the direction of the dip vector ap for each figure 6.

CLASSE DESCRIPTION OF JOINT SET NUMBERS

N1 Few discontinuities or occasional random discontinuities

a One main set
N2
b One main set plus random discontinuities
a Two main sets
N3
b Two main sets plus random discontinuities
a Three or more main sets
N4
b Three or more main sets and scattered discontinuities Figure 6 – Sketches of selected OR orientation classes from Table 11
N5 and stereogram data from figure 5
Abundant discontinuities in no discernible pattern Stereogram and plot of plane pole for the joint set considered
Table 10 – Classes and descriptions for joint set numbers (upper hemisphere) at left
Explanatory block diagram at right

14 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

SPACING (cm) DESCRIPTION

natural joints and included in the RQD cal-
CLASS culation as a conservative measure.
ES 1 > 200 Very widely spaced discontinuities • Discontinuities substantially parallel to the
core centreline to be ignored in the count.
ES 2 60 à 200 Widely spaced discontinuities
• RQD must be determined promptly after
ES 3 20 à 60 Moderately spaced discontinuities
the core is recovered in order to forestall
ES 4 6 à 20 Closely spaced discontinuities any subsequent changes in the material
ES 5 <6 Very closely spaced discontinuities due to swelling, stress release or drying;
the recommendation at the beginning of
Table 12 – Joint spacing classes within same joint set section 3 above to systematically photo-
graph all cores is critically important here.
Photographs taken at the time of the RQD
measurements of 2D trace lengths on sur- • Core diameter greater than 50mm in count allow any changes to the cores over
faces (outcrops, adit walls), but these data massive and poorly jointed rock mass; in time to be identified.
are often biased. Furthermore, 2D trace rock mass of a kind which is inherently join-
data can only be extended to actual 3D ted or incorporates planes of weakness RQD classes and assessments of the ove-
joint persistence with the aid of a model. rall quality of the rock (see para. 4.1.1) that
(schists, mudstones, marls, bedded limes-
can be derived from the D. Deere
Appendix 8 reviews recent developments tones, etc.), larger borehole sizes are
approach are presented in Table 13.
in these two areas. recommended, such as 85mm.
• Length of intact core pieces measured 3.2.3.2 – Interval between discontinui-
3.2.3 – Overall joint along the core centreline. ties (ID index)
density indexes The ID index is defined as the mean of the
• Core recovery index R (percentage of reco-
vered core lengths per core run legth) to be intact rock length between successive dis-
The spatial density of a population of dis- continuities along a survey line whose
continuities is defined as the mean joint 90-100%; the RQD index for R values lower
length and orientation must be recorded.
area per unit volume of rock. This definition than 90% is not considered meaningful.
The counts must be made in several direc-
encompasses both joint spacing and joint • Only natural discontinuities must be tions, carefully chosen with reference to the
persistence and cannot be determined considered; breakage caused by drilling, characteristic joint strikes and tunnel orien-
directly. It is usually only possible to esti- core handling and placing cores in the tation (this is also valid for the RQD
mate joint density by counting lengths of boxes should be ignored (this is particularly approach).
massive rock blocks between discontinui-
important in inclined boreholes which are The joint density classes obtained from the
ties encountered along a survey line in
more susceptible to core breakage); ID index are listed in Table 14
boreholes, outcrops or adit walls.
doubtful cases should be considered as
Joint data (without any morphogenetic
breakdown) is reduced for interpretation
by means of various indexes and statistical
manipulations. CLASS RQD % DESCRIPTION
(After D. Deere)
3.2.3.1 – Rock Quality Designation RQD 1 90 to 100 Excellent
RQD was defined by D. Deere (1963) as RQD 2 75 to 90 Good
the cumulative length of intact core pieces
longer than 4 inches divided by the total RQD 3 50 to 75 Fair
length of the core run not more than 1.50m RQD 4 25 to 50 Poor
long, expressed as a percentage. Core size RQD 5 0 to 25 Very poor
must be least NX (21/8 inches or 54,7 mm),
drilled with a double-tube core barrel, with Table 13 – Overall rock quality classes by the RQD method
* *after D. Deere
substantially 100% recovery.
In order not to penalise ground not yiel-
ding exactly 100% recovery and systema- CLASS ID INDEX (CM) DESCRIPTION
tise RQD records for the purposes of com-
parison, AFTES recommends calculating ID 1 > 200 Very low density
RQD from 1-metre core runs:
ID 2 60-200 Low density
RQD, expressed as a percentage, is the
ID 3 20-60 Moderate density
total length of pieces of intact core longer
than 10 cm divided by core run of 1-metre ID 4 6-20 High density
length. ID 5 <6 Very high density
In order to keep RQD meaningful, the fol- Table 14 – Joint density classes as determined along survey line
lowing coring conditions are critical:

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 15

 Caracterisation of rock masses useful for the design and the construction of underground structures

3.2.3.3 – Remarks on use of indexes

CLASS ID INDEX (CM) DESCRIPTION
It must not be forgotten that the determina-
FD 1 FD < 1 Very low density tion of the above indexes from borehole
cores requires the coring conditions stipula-
FD 2 1 < FD < 2 Low density ted in para. 3.2.3.1 always to be adhered to.
FD 3 2 < FD < 5 Moderate density Attention is also drawn to the fact that
FD 4 5 < FD < 15 High density RQD, the most commonly measured para-
FD 5 FD > 15 Very high density meter, is a quality index for the rock mass
Table 15 – Joint frequency classes as determined along survey line
but yields very little information on joint
density. For example, an RQD of 100 may
result from a core run not encountering a
discontinuity or exhibiting a discontinuity
every 11cm. AFTES therefore recommends
using the ID index to characterise joint
density.
It is also strongly recommended to repre-
sent the variations of these indexes along
the survey line in the form of diagrams. For
borehole measurements, it is essential to
accompany the complete core diagram
showing core piece lengths against depth
with the FD diagram (expressed in m-1)
and the RQD diagram (figure 8). This arran-
gement facilitates analysis of the rock mass
discontinuities, and the different methods
Figure 7 – Histogram and cumulative frequency curve for lengths off pieces of core (after C. Louis 1974)
become mutually complementary.
Homogeneous zones can be identified, or
It is also recommended (figure 7) image of the joint density. This curve can contrasts between zones with more or less
be used to quantify the mean joint index high density of discontinuities. The discon-
• To draw the histograms of lengths for
represented by the median, and dispersion tinuity data can also be compared with
each survey line and calculate standard
deviation S and coefficient of variation CV represented for example by the 25% and data from downhole logging performed in
75% quartiles. the same borehole.
CV = S/ID
The joint frequencies FD can also be calcu- It might also be useful to quantify these
• To plot the cumulative frequency curve of lated (Table 15); it is the number of joints variables by a moving mean process (e.g.
core lengths (equivalent to a grain size dis- per metre of drilled core, the inverse value RQD and FD calculated in one-metre steps
tribution curve) in order to have a complete of the ID index. in a 4m window moved down the borehole

Figure 8 – Characterisation of the

discontinuity network in a limes-
tone rock mass from vertical
cored borehole (Serratrice &
Durville 1997)

16 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

in 2m increments) or modify the values of 3.3.1 – Deformation parameters 3.3.2 – Shear strength parameters
the parameters used to calculate indexes
The behaviour of a discontinuity during a
in specific zones in the borehole such as 3.3.1.1 – Normal stiffness
shear test (French standard XP P 94-424) is
crushed zones which can conventionally be Uniaxial compression tests on joints orien- governed by the nature of the joint walls
considered as consisting of 1cm fragments. ted perpendicular to the direction of load but more importantly, by their surface
application always trace a hyperbolic curve conditions. In particular, joint wall rough-
3.3 – MECHANICAL PARAMETERS
of normal stress σn versus normal displace- ness, interlocking and weathering play a
OF DISCONTINUITIES
ment Un with an asymptote representing primordial role.
We are interested here in discontinuities the maximum limit of joint closure Umax.
without infilling material, otherwise mecha- Total joint opening can be obtained for a In the ideal case of a planar and smooth
nical behaviour would be governed by the non zero tensile stress a, if there are rock discontinuity, i.e. with no asperities, shear
behaviour of the infilling material, which bridges across, or filling material in, the behaviour is entirely controlled by wall fric-
should then be studied in itself. joint (Figure 9). tion. Shear strength is usually expressed by
The mechanical characteristics of disconti- the Coulomb criterion:
The slope of this curve gives the normal
nuities are usually obtained from laboratory stiffness Kn, defined as τ = σn x tan φb
tests; in situ tests are much less common in which φb is the friction angle for a planar
because they are more difficult to perform Kn = δσn/δUn
joint or basic friction angle, chiefly depen-
and are more costly. Uniaxial compression The value of Kn is dependent on the normal dent on the petrographic composition and
tests and shear tests with normal load stress and can be expressed in terms of degree of weathering of the joint walls.
(French standard NF P94-424) are perfor- parameters α, Umax and Kni (initial normal
med to characterise the behaviour of the stiffness) characterising the mechanical Natural discontinuities generally have walls
discontinuities. behaviour of the joint subjected to uniaxial which are very irregular with abundant
The following leading parameters charac- compressive load, determined by fitting a asperities of varied shape and size, repre-
terising the mechanical behaviour of dis- curve on the test results, although not com- senting different scales of roughness. Their
continuities can be obtained by analysing monly done. shear behaviour reveals three fundamental
these laboratory test results : parameters (figure 10):
3.3.1.2 – Tangential stiffness • Peak shear strength, defined by the maxi-
• deformation parameters: normal and tan-
gential stiffness, Similarly, the shear test is used to define mum shear stress τp, at which the asperities
the tangential stiffness Ks as the slope of shear.
• shear strength defined by peak and resi-
the curve of tangential stress τ vs tan- • Residual shear strength τr, characteristic
dual friction angles and apparent cohesion,
gential displacement U s before failure of the friction of the joint walls which come
• a geometric parameter, dilatancy, a mea- (figure 9): into contact with each other after the aspe-
sure of the deformation in the normal
direction accompanying tangential defor- Ks = δτ/δUs rities have sheared.
mation during shear.

Figure 10 – Shear behaviour of a natural discontinuity

a) tangential stress τ vs tangential strain Us
Figure 9 – Joint deformability parameters: normal stiffness Kn and tangential stiffness Ks b) Normal strain Un vs tangential strain Us

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 17

 Caracterisation of rock masses useful for the design and the construction of underground structures

• Dilatancy represented by the displace- straight line. For a limited normal stress It should be recognised that this approach
ment of the joint walls in the direction nor- range, this curve can be approximated by a is confined to discontinuities with thin or no
mal to the joint plane. It is characterised by straight line: infilling material. When there is a sufficient
the dilatancy angle i (angle of slope of the thickness of infilling material for shearing to
τpeak = Ca + σn x tan φpeak
dilatancy curve of normal displacement Un occur wholly within the infilling material,
vs tangential displacement Us). This angle Ca is an apparent cohesion which does not shear characteristics will be those of the
reaches a maximum value ip at the inflec- express an intrinsic property of the joint infilling material, which must be investiga-
tion point on the dilatancy curve. This point wall material but the influence of irregulari- ted specifically.
corresponds to the peak shear stength τp ties in the walls on shear behaviour.
reflecting, for a given level of normal stress, At very low normal stresses, apparent
the shearing of the sharpest asperities. cohesion Ca is close to zero, and φpeak is
Beyond this point, dilatancy continues with close to φr + ip.
a lower angle, governed by the inclination
of the stronger asperities with a wider base At high normal stresses, apparent cohesion
and flatter angles. Ca is high and peak friction angle φpeak
tends towards φr.
Compared to a planar smooth joint, dila-
tancy leads to an increase in peak strength. In practice, laboratory tests are not easy to
It is dependent on joint wall roughness and interpret and determination of peak joint
weathering, and also how the walls inter- strength characteristics involves many diffi-
lock and the direction of shear. culties arising from scatter in the data and
scale effect.
The joint shear failure criterion is represen-
ted by two curves, characterising the peak Using experimental data, Barton (1973)
and residual strength (figure 11). proposed a semi-empirical failure criterion
in which peak strength depends on a dila-
tancy angle i allowing for the joint wall
roughness (JRC), joint strength (JCS) and
normal stress applied to the joint:
τpeak = σn x tan (φb + i)
= σn + tan [φb + JRC x log10 (JCS/σn)] Figure 12 – Standard joint wall roughness profiles
(after Barton & Choubey 1977)
in which φb is the basic friction angle which
differs from the residual friction angle φr by
a few degrees.
3.3.3 – Hydraulic parameters
JRC is the Joint Roughness Coefficient, a
Figure 11 – Failure criterion for a natural discontinuity
dimensionless coefficient relating to joint While the mechanical behaviour of rock
wall roughness and size. It can be estima- joints is mainly controlled by joint wall com-
ted by comparing joint roughness profiles position, weathering, roughness and nor-
Residual strength of discontinuities is not mal stress, other external factors affect
in the direction of shear with Barton's stan-
greatly influenced by scale effect and the behaviour: thickness, composition and
dard profiles, ranked in ascending order
failure criterion is readily obtained by labo- moisture content of infilling material, pre-
from 0 for a flat smooth discontinuity to 20
ratory testing in the form of a standard sence of water in joints likely to induce
for a wavy rough discontinuity (figure 12).
Coulomb law characterised by a residual pore pressures modifying normal stress,
JRC also varies with the joint deformability
friction angle φr which differs from the basic and boundary conditions affecting the
wall displacement: the more asperities are
friction angle φb by no more than a few magnitude of displacements.
sheared, the lower the value of JRC.
degrees, and a residual cohesion, usually Fracture fluid flow is a highly complex sub-
minimal or zero, which is always considered RCS is the Joint Compressive Strength, fre- ject. Experiments have shown it not isotro-
to be 0: quently estimated indirectly in situ with the pic but occurs preferentially along channels
τr = σn x tan φ Schmidt hammer (Appendix 9). whose geometry depends, of course, on
The peak shear strength curve has a pro- σn is the normal stress applied to the dis- the aperture of the discontinuity, but also
continuity. on wall roughness and surface of contact
gressive shape reflecting the non-linear
between walls, applied normal and tan-
relationship between shear strength τ and At very low normal stresses (JCS/σn ≥100),
gential stresses, and tangential joint displa-
normal stress σn. The curve is steep at low the equation gives unrealistic values and
cements, plus of course the possible pre-
normal stresses, reflecting the influence of Barton suggests using the simplified form: sence of infilling material.
the sharpest asperities, which are the cause
τ = σn x tan 70° Various more or less simplified approaches
of severe dilatancy. As normal stress
increases, more and more asperities fail, However, the determination of a represen- can be used to estimate the flow rate Q of
dilatancy becomes less and the (τ, σn) curve tative value of JRC for three-dimensional a fluid circulating in a discontinuity (see
flattens and progressively becomes a joint wall roughness is not always a simple appendix for details). For a planar and
matter, even on laboratory size samples. smooth discontinuities, the flow is gene-

18 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

rally assumed to be proportional to the

cube of the discontinuty aperture
(Appendix 10).
But even allowing for roughness, it must
never be forgotten that measured flows Figure 13 – Influence of
direction on the characterisation
often differ substantially from rates calcula- of discontinuities in finely
ted by these methods. bedded formations

4 – CHARACTERISTICS OF
ROCK MASS

4.1 – IDENTIFICATION of explosives should be ignored as far as With weathering properly so called, des-
PARAMETERS possible. criptive terms, conforming to those
recommended by ISRM (AM = W), appear
If the rock mass displays strongly polari-
4.1.1 – RQD in Table 16. They apply predominantly to
sed discontinuities, the same reservations
crystalline rocks.
RQD determined from jointing (see para. can be made as in the case of boreholes
3.2.3.1) was originally considered as an as to the representativeness of the dis- 4.1.3 – Rock mass continuity
index of rock mass quality determined by continuities recorded and the RQD calcu- index ICM
counting discontinuities in borehole cores. lated with reference to the direction of
the survey line. Using the same procedure as described in
If the discontinuities are all oriented more para. 2.1.10 for the rock matrix, a rock
or less uniformly in all three dimensions 4.1.2 – Degree of alteration mass continuity (or quality) index ICM can
('isotropic' jointing), RQD can be taken as be defined as the ratio between P wave
independent of the direction of the bore- The degree of alteration of a rock mass is velocity as measured over a base length L
described by breaking it down into altera- (Vp M) and the velocity measured on a
hole and can effectively be considered as a
tion zones for the different geological for- sample (Vp):
overall index of the quality of the rock
mations present. A distinction is made bet-
mass. ICM = VpM/Vp
ween weathering proper and hydrothermal
But if the distribution of discontinuities is alteration occurring at depth (frequently This concept of a rock mass continuity
strongly polarised (as in finely bedded linked to contemporary or more ancient index ICM makes it possible to estimate the
rocks, schists, slates, etc.), the value of the volcanism). Alteration of the rock mass as a impact of the scale effect and deterioration
RQD index will differ widely with different whole is classified as the sum of the wea- in the mechanical properties the rock mass
directions of drilling (figure 13). The RQD thering of the rock matrix and of the major compared to results from laboratory
from a single borehole therefore will give joints. samples (rock matrix).
only a 'snapshot' of the jointing in a given
direction rather than a representation of
the overall jointing of the rock mass.
AFTES DESCRIPTION OF THE ROCK MASS
Because of this, AFTES recommends that CLASSES
RQD should be determined from several
boreholes drilled in different directions to AM1a Sound rock
intersect all joint sets, especially those
AM1b Poorly weathered rock
which may be unfavourable for the plan- Weathering confined to surfaces of main discontinuities; rock sound in the mass
ned underground structure.
AM2 Slightly weathered rock
RQD, originally defined by its author on
Little weathering of rock in the mass but well developed in discontinuities
the basis of counting discontinuities
found in borehole cores, can also be AM3 Moderately weathered rock
determined by counting them on expo- Weathering clearly visible in whole rock mass but material not friable
sed rock surfaces:
AM4 Well weathered rock
• Natural outcrops: the count proceeds Severe weathering in the mass
along one or more lines intersecting the
AM5 Completely weathered rock
network of fractures such that the values Texture and large fractures still visible
obtained is truly representative of the
homogeneous blocks of rock as defined AM6 Completely decomposed rock
elsewhere. Texture and fractures unrecognisable
Residual soil - Undisturbed
• Quarry faces, trench sides, adit walls: in
these cases, cracking caused by the use Table 16 – Rock mass weathering classes and descriptions

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 19

 Caracterisation of rock masses useful for the design and the construction of underground structures

Whatever the absolute value of Vp, if VpM is • 'Direct' methods consisting of measuring 4.2.1.2 – Direct measurement
equal to Vp, this means that the rock mass, deformations on parts of the rock mass The main in situ tests for measuring rock
at the scale L at which VpM was measured, under changing states of stress. The mass deformability are:
displays the same properties as the sample changes may be brought about by specific
and is unaffected by discontinuities or voids load tests (in situ tests) or by construction of • Rigid plate loading test, quite wides-
which would reduce P wave velocity. the tunnel (back analysis). pread and routine. It characterises the
deformability of the rock mass through
If however, VpM is less than Vp, the lower 4.2.1.1 – Indirect (geophysical) methods deformation modulus E determined from
velocity in the rock mass than on the labo- the tangent to the envelope curve of the
ratory samples can be attributed to discon- P (longitudinal compression) wave and S
'load-displacement' curves under increa-
tinuities and voids in the rock mass over (transversal shear) wave travel time is mea-
sing loading cycles (cf. Appendix 11).
base length L at which VpM was measured. sured over a known distance between the
seismic source and pick-ups. Emitter and With usual plate sizes (0.28m to 0.60m),
Rock mass continuity classes, at the scale pick-ups can be arranged in various ways: this test yields rock mass deformability
of L over which VpM was measured, are lis- values at the scale of a few cubic metres of
ted in Table 17. • In separate boreholes (seismic cross hole
test) rock, provided the pressure is high enough
Base length L is usually the same as stan- to penetrate beyond the decompressed
dard seismic refraction test base lengths • Source in a borehole and pick-up at surface zone.
(60m, 120m, 240m) but rock mass conti- ground level (seismic up-hole test)
• Borehole dilatometer test (French stan-
nuity can be measured over shorter base • Pick-up in the borehole and source at sur- dard P 94-443). The instrument consists of
lengths, based on borehole microseismics face (seismic downhole test). a deformable cylindrical cell applying a
or adit wall seismic tests. All arrangements measure compression controlled radial pressure to the borehole
Of course, base length L must always be wave Vp and shear wave Vs velocities to walls, and several strainmeters directly
stated alongside the relevant ICM value. derive the 'dynamic' deformation modulus measuring the radial deformation of the
and 'dynamic' Poisson's ratio of the rock borehole wall under the applied pressure.
Note. ICM may be greater than 100%, for
mass (Ed and νd respectively) through the The E modulus of thr rock mass is calcula-
example when the rock matrix contains
following equations: ted with the following equation in which, in
cracks or microcracks closed tight by the
confining pressure in the rock mass but Ed = ρ[Vp2(1 + νd) x (1 – 2νd)]/(1 - νd) the absence of specific data, the Poisson's
which might open when coring releases νd = [0.5 – (Vs/Vp)2]/[1 – (Vs/Vp)2] ratio n is frequently taken as 0.25:
these stresses. This is a not uncommon
(ρ is density, see para. 2.1.4). ∆σr/∆εr = E/2(1 + ν)
occurrence in some schistose rocks.
Computing Ed and νd by geophysics means ∆ε is the change in radial strain produced
measuring both Vp and Vs. by the change in applied stress ∆σ on the
4.2 – MECHANICAL borehole wall.
PARAMETERS If only the compression wave velocity Vp
has been recorded, the Ed modulus can still The strainmeters measuring the radial
be obtained by taking an assumed value deformation at the borehole wall must dis-
4.2.1 – Rock mass deformability, play sufficient resolution to measure the
rock mass deformation modulus EMas for Vs (usually 0.25 or 0.30).
usually high moduli encountered in rock.
Because of the discontinuities, rock mass Other methods based on seismic velocities They are arranged in different directions at
deformation at prototype scale is generally in the rock mass can yield estimates of the different places along the length of the dila-
much greater than for the intact rock matrix corresponding moduli, through experi- tometer (3 pairs at 120° or 4 pairs at 90° in
as determined from small laboratory mental correlations with past construction different models). This arrangement shows
samples. sites (Schneider's 'Petite Sismique' up any anisotropy in rock deformation.
method, SCARABEE method).
Depending on rock mass volume and the The dilatometer can also be used for creep
loads applied to it, deformability may be It must be realised that the term 'dyna- tests in which the applied pressure is held
apprehended through two classes of in situ mic' here in fact refers to very small strains constant over time and time-dependent
investigations: (10-7 < ε < 10-5) under very small loads.
displacements are recorded.
• 'Indirect' geophysical methods, based • Borehole pressuremeter test (standard P
primarily on wave velocities. 94-110-1 & 2) also measures mass defor-
mability by means of a deformable cylindri-
CLASS ICM DESCRIPTION cal cell applying an increasing pressure to
the borehole walls. The slope of the 'pres-
ICM 1 > 90 % Very high continuity suremeter curve' showing the change in
pressuremeter cell volume vs applied pres-
ICM 2 90 % to 75 % High continuity
Table 17 – Rock mass
sure is used to calculate a shear modulus G
ICM 3 75 % to 50 % Moderate continuity continuity classes at L scale (Menard pressuremeter modulus EM).
ICM 4 50 % to 25 % Low continuity However, the characteristics of the instru-
ICM 5 < 25 % Very low continuity ment mean that the pressuremeter test is
unsuitable for rock mass determinations,

20 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

even if it is still widely used. The rea- behaviour (as in certain rocks such as
son is that, with moduli of a few hun- evaporates, marls, etc.).
dred MPa (a value which is abun- • The elastic limit may be exceeded
dantly exceeded in rock), the moduli with plastic zones developing around
from pressuremeter tests become the opening.
increasingly underestimated whereas
actual moduli increase. The pressure- • Time-dependent deformations may
meter test must only be used for be linked to consolidation processes
soils and some 'soft' materials (chalk, subsequent to changes in flow pat-
marl) on the borderline between soil terns, with the opening acting as a
and rock. When dealing with the drain, or to the original pore pressure
usual types of rock, the pressureme- patterns gradually re-establishing after
ter should not be reckoned among the disturbance caused by excavation.
the panoply of relevant test Photo 7 – Rigid plate loading test.Volcanic agglomerate, These causes may be concomitant,
methods. Takamaka, Reunion Is. and this aggravates the difficulties of cor-
rectly interpreting the observed time
4.2.1.3 – Measurement on actual dependent behaviour.
structures and estimation of When checking the design of very high
deformability by back analysis head water pressure tunnels, the concrete At present, the most widely used simpli-
lining subjected to the high water pressure fying approach to the understanding of
These are probably the most effective
is instrumented (to measure diameters and time dependent rock mass behaviour is to
methods for finding the large scale defor-
stresses). This type of 'chamber' test also consider the rock mass deformation modu-
mability of a rock mass and the anisotropy
provides a check on rock watertightness. lus as a decreasing time function:
parameters governing it.
EMas(t) = EMas0/[1 + Φ(t)]
On actual structures (generally exploratory 4.2.1.4 – Classification of rock mass
adits driven prior to the full size construc- deformability in which
tion stage), the measurements most com- Rock mass deformability classes based on - EMas0 is the instantaneous rock mass defor-
monly performed are the following: the rock mass deformation modulus EMas mation modulus
• displacements of the adit wall (conver- are listed in Table 18. - E Mas(t) is the rock mass deformation
gence) modulus at time t under a load that has not
4.2.1.5 – Time-dependent effects –
• displacements of points within the rock long term modulus changed since time t = 0.
(displacements relative to fixed or moving Φ(t) is a monotonous function increasing
reference points) by means of borehole The construction of an underground struc-
ture always causes deformations in the sur- from Φ(0) = 0 to Φ(∞) = α.
extensometers around the adit
rounding rock due to changes in the stress In low- to moderate-strength rock, α = 1 is
• angular changes between studs fixed to field around the opening. In many cases, often suggested even if it does not always
the rock (inclinometers or deflectometers). rock deformations are accompanied by appear justified by experimental and
Back analysis, generally with 2D or 3D effects which appear over time and defor- other data.
computer models (finite element and simi- mations increase asymptotically towards In stronger rocks, α = 0.3 to 0.5 are also fre-
lar models) allows the engineer to work what is generally called the "long term quently proposed without any justification
back from the known stress state to the state." for the choice.
most plausible rock moduli under the As discussed in para. 2.2.2, there may be
conditions of the completed structures. In A less doubtful choice of these values
several causes of time dependent rock would require long-term instrumental data
the case of anisotropy in the moduli, which mass behaviour:
usually accompanies anisotropic states of from completed structures. At the present
stress, back analysis is more difficult. • Rheological behaviour specific to the time, little feedback is available and
rock mass, viscoelastic or viscoelasticplastic records cover only a limited number of
years.

4.2.2 – Rock mass limit strength

CLASS ROCK MASS DEFORMATION DESCRIPTION
MODULUS EMas (GPa)
Mechanical properties are strongly influen-
ced by the geometrical dimensions of the
DM 1 > 30 Very low deformability volumes of rock involved, in the general
DM 2 10 to 30 Low deformability sense of these properties becoming worse
as the rock volume grows (scale effect). In
DM 3 3 to 10 Moderate deformability situ rock mechanics tests to failure such as
DM 4 1 to 3 Fair deformability shear tests and hydraulic fracturing involve
DM 5 0.1 to 1 High deformability considerable effort and still only concern
limited volumes of rock.
DM 6 < 0.1 Extremely high deformability
Table 18 – Rock mass deformability classes

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 21

 Caracterisation of rock masses useful for the design and the construction of underground structures

There are not really any tests for characte- applicable. The hydrogeological functio- zometric testing may be indicated, using
rising the mechanical strength of a rock ning of the system is then tentatively pressure cells; open well piezometers
mass. Only an empirical approach based modelled, with rough estimates, for each should not be used.
on feedback from past construction jobs is aquifer crossed by the project, of Piezometry is usually subject to seasonal
possible. This leads to modifying the crite- a) The type of permeability concerned, fluctuations and it is strongly recommen-
ria for samples (see para. 2.24) by downs- which can be classified under five hea- ded that piezometric monitoring of each
caling the characteristic parameters when dings: aquifer identified should commence at the
extrapolating from the intact rock matrix to very earliest stage of the design process. It
the large scale rock mass (see 5.3.4). 1. Granular material (sand and gravel);
will often be necessary to have several
2. Jointed rock (granite, gneiss, basalt, etc.) years' records before being sure of the
4.3 – HYDROGEOLOGICAL with water circulating only through the dis- amplitude of the piezometric fluctuations
CONDITIONS continuities; to be expected and designed for. In most
3. Double porosity ground in which water cases, only continuous records will show up
Ground water is the cause of many difficul- sometimes short-lived transients which
circulates both through discontinuities and
ties encountered in underground enginee- may have a serious impact on the project
the porous rock matrix (chalk, sandstone)
ring: (karstic aquifers, tidal river reaches, etc.).
or weathered rock matrix (severely weathe-
• Flowing water slows down excavation red granite); The designer must also assess the risk of
work the water table rising, especially in urban
4. Karstic rock (limestone, gypsum) in
• Water pressures may destabilise the tun- areas, due to local abstractions being
which most of the water circulates through
nel walls or lead to fearsome squeezing unexpectedly interrupted. Lastly, know-
randomly distributed voids of various sizes;
ground into the tunnel ledge of changes in the peizometry and
5. Fault zones with breccia infill frequently flow rates leads indirectly to certain aquifer
• Dewatering may have severe environ- acting as drains within fractured rock- hydrodynamic parameters.
mental consequences: flow depletion from masses.
springs and wells, subsidence due to Hydraulic head classes are listed in Table 19.
b) Boundary conditions, i.e.
groundwater lowering.
- sources (rainfall, infiltration, river, lake, INITIAL HYDRAULIC
Basically, flow rate Q through a section S of
sea, etc.) HEAD H
the rock mass is related to permeability K CLASS (in metres above tun- DESCRIPTION
and hydraulic gradient i by Darcy's law: - flow rate at point of outflow nel invert
Q = K.S.i - watertight boundaries.
Flow is associated with body forces propor- This first step involves a desk search and H0 Lower than invert Zero head
tional to the hydraulic gradient i. inspection of a project survey plot; it uses H1 <5 Low head
data from the geological model and
Characterisation of the hydrogeological H2 5 to 20 Moderate head
demands considerable engineering expe-
conditions in a rock mass therefore pro- H3 20 to 100 High head
rience. It should not overlook local sources
ceeds in three steps:
of information from groundwater users, H4 > 100 Very high head
1. Identify aquifers and how they function. amateur kart environment, etc. In karstic
2. Measure hydraulic head H on the tunnel. settings, it is important to know whether Table 19 – Hydraulic head classes
the karsts are active or fossil karsts more or
3. Measure rock mass permeability KM. less filled in. As a general rule, every cored borehole
In practice, these three steps do not neces- sunk to investigate a tunnel site should be
sarily proceed in the order described 4.3.2 – Measurement of initial fitted out as a piezometer, in view of the
because it may not be possible to eluci- piezometric conditions importance of this parameter.
date the existence of certain aquifers and Knowledge of the pre-construction piezo-
their functioning until exploratory works metry in the rock mass at the project site is 4.3.3 – Measurement of rock mass
have been undertaken. It is also important a critically important factor in good design. permeability KM
to realise that such investigations must The project may be a very long tunnel and
cover the whole aquifer system affected by pre-construction piezometry must be Determining the permeability of the rock
the tunnel, not only the part through which determined along its whole length. It mass calls for interpretation of the results
the tunnel passes, as is the case with inves- seems unnecessary to stress the risks ari- of hydraulic tests, which must be chosen
tigations for mechanical parameters. sing from ignorance of the initial piezome- to suit the type of aquifer concerned.
tric conditions (even over a short length of Available tests are as follows.
4.3.1 – Identification of aquifers tunnel) both during construction and sub-
sequent operation of the structure. • Localised tests in boreholes:
The extent of the aquifer system liable to
be affected by the underground structure Where more than one aquifer and different - either steady-state tests where permeabi-
is determined with the aid of the geologi- pressure heads are suspected and in zones lity is moderate to high, such as the stan-
cal model mentioned in section 1 above, where the relief may be an influencing fac- dard Lefranc test for soils and Lugeon test
identifying the main individual aquifers if tor (hillsides, valley bottoms, etc.) local pie- in jointed rock (water is injected at 1 MPa

22 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

pressure); the Lugeon test measures appa- • hydraulic head, by measuring pressures 4.3.5 – Other parameters
rent permeability and supplements the between packers,
In addition to the H and K parameters for
jointing data obtained from cores and • borehole inflow/outflow, by continuous each aquifer identified, other parameters
mechanical behaviour; measurement with a miniature flowmeter. may be of use in characterising the rock
- or transient-state tests, causing instanta- When dealing with deep-lying mine wor- mass, in particular:
neous changes in the open (slug test) or kings, tunnels, mined storage chambers • Storage coefficient S, representing the
closed (pulse test) test interval, or a combi- and other deep structures, understanding capacity of the rock mass to store water.
nation of them (drill stem test); these more the hydrogeology is complicated by the This must be investigated for modelling
complex tests are suitable for low permea- possibility of exploratory drilling causing transient flow conditions; it can be derived
bility conditions (K < 10-7 m/s). unwanted interconnections between sepa- from pumping-out test data and may range
• Larger-scale tests, such as: rate aquifers. This calls for more sophistica- from 10-5 (10 cm3 of water released when
- pumping-out tests with measurement of ted techniques as used in the mining and the hydraulic head in a 1 m3 volume of
far-field water table drawdown in suitably oil industries, such as those described in saturated ground is lowered by 1m) to 0.1
located boreholes; the AFTES Working Group 24 to 0.15 in clean sands.
Recommendations. They accurately locate
- measurement of tunnel drainage (the tun- aquifers through the use of fluid logs and • Groundwater temperature, pH and che-
nel may be divided up into separate multilayer tests, and make it possible to mistry (and sometimes isotopes). These
lengths for this purpose). test them selectively with packers and parameters serve more as indicators of
When conducting these tests, it is important probes. where the water comes from and help
to ensure that the disturbance induced by understand the hydrogeological functio-
the test is of the same order of magnitude as ning of the aquifer system; and they make
the disturbance that will be caused by the it possible to assess how aggressive the
tunnel, so as not to disrupt the environment. water will be for tunnel support and linings.
In all cases, upscaling measured data to the
whole aquifer calls for great caution. 4.4 – INITIAL STATE OF
STRESS IN ROCK MASS
Rock mass permeability classes are listed in
Table 20. The initial stress state is a determining fac-
tor in the response of the rock mass to
ROCK MASS
CLASS PERMEABILITY DESCRIPTION excavation: the convergence pattern on a
KM (m/s) tunnel section, the location and extent of
Photo 8 – Water flow into Saint Guillaume II tunnel from zones where the limit strength of the rock
K1 < 10-8 Low permeability Grange Pellorce fault, France may be reached during tunnel driving, are
K2 10-8 à 10-6 Moderate permeability
all strongly dependent on the initial stress
K3 10-6 à 10-4 High permeability state, and it is vital to consider it at the
4.3.4 – Gas
K4 > 10 -4
Very high permeability design stage.
Methane (CH4), nitrogen (N2), hydrogen
K5 Pratically infinite Karst permeability
sulphide (H 2S), carbon monoxide and 4.4.1 – Initial state of stress and
Table 20 – Rock mass permeability classes dioxide (CO and CO2), radon 220 or 222 approximations
(Rn) and other gases may be present in the
free state or dissolved in the ground water Computation and modelling in the design
Permeability in jointed rock is very often
within certain sedimentary formations (car- stage make it possible to investigate and
anisotropic: it may typically be ten times
bonaceous, carbonate, argillaceous and analyse the impact of the initial state of
greater parallel to the bedding or cleavage
saline rocks) or igneous formations (e.g. stress.
than in the perpendicular direction.
Anisotropy may also depend on the orien- granite). When such formations host an The state of stress is represented everyw-
tation of the principal stresses. Equivalent underground opening, the gases they here by a tensor whose principal compo-
permeability for the whole rock mass is contain tend to migrate towards the exca- nents are the σ1 (major), σ2 (intermediate)
given by a tensor; for classification pur- vation, creating a risk of explosion, poiso- and σ3 (minor) stresses.
poses, the highest permeability coefficient ning, suffocation or disease (cancer and
In the absence of data, it is commonly assu-
is used, stating the direction in which it other occupational illnesses), not only
med that the vertical is the principal direc-
applies; the anisotropy ratio Kmax/Kmin is during excavation, but equally during the
tion and that the vertical stress is equal to
also used. service life of the structure. While such risks
the "weight of overlying ground," i.e.
are more the province of health and safety,
When running borehole tests, it is recom- σv = γ x z
they must be addressed at the tunnel and
mended to proceed in contiguous 5m to
ventilation system design stage and must These two assumptions are valid in subho-
10m stages and plot a permeability log
therefore be accorded special attention rizontal sedimentary formations but are not
which can be usefully compared to the
when proceeding with the geotechnical generally valid in mountain areas where the
jointing log.
characterisation of the ground to host the relief and tectonics introduce considerable
Other useful data that can be logged structure when such gases may be present distortions, especially under mountainsides
concerns: within it (coal measures for example). and at valley bottoms.

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 23

 Caracterisation of rock masses useful for the design and the construction of underground structures

In sedimentary basins, a third assumption is The estimated stress range and its impact feasibility, it may be decided to undertake
frequently made, that the horizontal princi- on design may subsequently justify perfor- field testing at an earlier project planning
pal stresses σh and σH are identical and ming in situ field tests. If severe horizontal stage.
equal to a constant fraction of σv: anisotropy is considered possible at this
σh = σH = K0 x σv stage, data on the azimuth of sH may lead 4.4.3 – Commentary on field test
to the orientation of the underground ope- methods
This assumption may be very far from rea- ning being optimised (provided such free- When measuring in situ stresses, it is stron-
lity since the horizontal stresses are rarely dom is possible, in view of the purpose of gly recommended to plan an abundant
isotropic and K0 frequently ranges from 0.5 the structure). number of mutually complementary tests,
to 2 or more 5.
At the detailed site investigation stage, the not only because of metrological difficul-
4.4.2 – Characterisation of stress available methods for stress measurement ties, but more importantly because of the
tensor are always considered, not without reason, often very sudden local variations in the
expensive, difficult to perform and interpret stress tensor; these variations may be due
When designing any underground struc- and above all extrapolate. On top of this, to lithological heterogeneities or the proxi-
ture, it is important to try to determine the one can never directly "measure" a state of mity of geological discontinuities, fracture
initial state of stress. Because of the difficul- stress; at best, one measures fluid pressures zones or even a free surface (favouring
ties involved in this, a step-by-step proce- considered as equivalent to the normal stress release). Such conditions make extra-
dure is usually followed, based firstly on component acting on a given surface, or polation of test data even more problema-
indirect approaches and then, if possible, strains caused by stress changes. Despite tical and it is vital to check the data during
on in situ test data which can be checked these difficulties, it is important to have ins- construction work.
during construction by specific observa- trumental data, especially when the presu-
tions (Table 21). The various test methods available have
med mean stress is of the same order of been described in Tunnels et Ouvrages
In the project planning stage when an magnitude as rock strength (cf. para. 4.4.4). Souterrains No. 123 (Briglia et al. 1994).
approximate estimate of stresses is suffi- Lastly, once construction has commenced, Only a few general recommendations will
cient, the designer focuses on indirect ana- the stress assumptions derived from the be given now, focusing on borehole
lyses, using published information and field tests have to be viewed alongside the methods.
conclusions that can be drawn from the actual response of the tunnel walls and
geological history and local topography of a) Methods based on stress release, by
data from monitoring instruments. overcoring, undercoring, or cutting a slot
the project area. This first step should pos-
tulate a stress range to be expected. If the initial state of stress is found to be a with a borehole slotter, usually in boreholes
determining factor in project design and that are differently oriented; data on rock
deformability is needed. The CSIRO,
PROJECT USBM and other instruments involved are
STAGE OBJECTIVES METHODS AND MEANS quite difficult to use properly and are
mainly suitable for poorly jointed rock, in
• Regional tectonic regime (compres- Published data: which they yield purely local information
sive, strike-slip, extensional) • Stress maps (on a decimetric scale); they can only be
Project • Orientation of major principal • Mechanisms at regional earthquake used in the elastic range.
stresses hypocentres*
planning • Local disruptions due to Quater- • Geological and topographic maps b) The hydraulic fracturing method mea-
nary geological processes (palaeo- • Geotechnical reports on existing sures the normal component of the stress
relief, glaciations, erosion, etc.) structures acting on a discontinuity, by means of an
• Influence of relief on state of stress • Palaeogeographic assumptions elevated water pressure on a section of
borehole. It has the advantage of involving
• Estimate of σH/σV ratio Data from deep boreholes a volume of rock several cubic metres in
• Orientation of major horizontal • Ovalisation of bore size, being feasible at great depths (in
stress σH • Disking in cores excess of 1000m) and needing no assump-
Design tions on rock behaviour. There are two
• Determination of complete tensor Borehole stress measurements
if possible • Overcoring and borehole slotter variants:
• Hydraulic fracturing and HTPF • "Standard" hydraulic fracturing, creating
Flat jack tests in adits artificial fractures perpendicular to the
minor principal stress s3 to determine its
Validation of design assumptions • Observation of tunnel walls magnitude and direction.
Construction
based on exploratory works • Interpretation of strain data • The hydraulic test on pre-existing frac-
tures (HTPF test) extends the scope to
*As an initial approximation, the P and T axes of focal mechanisms can be taken as s1 and s3 natural fractures with different orientations.
Provided enough tests are performed,
Table 21 – Stress estimation methods at different project stages HTPF is one of the most reliable means of
determining the complete stress tensor.

5
Strictly speaking, coefficient K0, more commonly used in soil mechanics, is the effective stress ratio

24 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

If an exploratory adit is available, stresses CLASS σc/σ0 RATIO DESCRIPTION OF STRESS STATES
can be measured at different points in the
adit wall by the flat jack method. It is only
suitable in rock having few or no joints that CN 1 >4 WEAK
has been substantially unaffected by exca- Rock matrix satisfactorily strong but support may be needed
vation. The only assumption needed is that because of jointing
rock response will be reversible, but calcu- CN 2 2à4 MODERATE
lation of the complete tensor relies on a Failure or plastic zones possible at tunnel walls
model of the rock mass around the adit.
Tests must be made over several straight CN 3 <2 STRONG
sections of the adit. Rock matrix strength manifestly insufficient

Table 22 – Preliminary classification of relative state of stress around a tunnel

4.4.4 – Classification of stress
states
thermal properties of the rock mass under At the design stage, a more detailed search
In the project planning stage, the designer pure conduction conditions (where there is is made for anomalies by systematically
can make his first assessment of the gene- no flowing fluid) and it is usually taken that measuring water temperature in explora-
ral stability of the unsupported opening on rock conductivity at depth is close to the tory boreholes (temperature-conductivity
the basis of the σc/σ0 ratio, with σc repre- value measured on samples. logs).
senting the uniaxial compressive strength
In contrast, the effect of anisotropy must During construction, temperature measure-
of the intact rock matrix (the most readily
be considered because conductivity may ments as tunnelling proceeds allow the
accessible parameter at this stage) and σ0
be up to twice as high in the direction parameters determined in the previous
being the value of the major stress in a parallel to the cleavage than in the perpen-
plane normal to the tunnel. phase to be checked and validated. Any
dicular direction. thermal anomalies detected in boreholes
On the basis of the σc/σ0 ratio, one can pro- A rock mass is naturally supplied with heat drilled ahead of the face are excellent war-
ceed to the general qualification of the ini- ning signs that the drive is nearing zones of
tial state of stress with respect to its conse- • from underneath, due to the geothermal
flowing underground water and geological
quences on the planned structure. flux Φ, which can be considered constant at
discontinuities.
the scale of civil engineering construction.
Table 22 lists stress state classes based on There is little precise local date on this point. With very large projects, temperature
the σc/σ0 ratio. It usually ranges from 50 to 80 mW/m2. In modelling with thermal models or suitably
It is absolutely essential to estimate the pure conduction, we have modified hydraulic models (Goy 1996) is
stresses by means of field tests when the Φ=Γxλ possible after precise determination of sur-
structure can be expected to lie in class CN2 face temperature, structural geology, rock
or CN3 conditions (assuming σ0 = γ x z). • from the surface: heat from sunlight conductivities and local geothermal flux
varies with time of day and season but such (pure conduction models are valid if there
4.5 – TEMPERATURE changes do not affect rock temperature is no significant water flow, Fabre 2001).
beyond about twenty metres’ depth; for
It is only necessary to predict natural tem- temperature modelling purposes, sunlight
peratures in the rock mass for some under- governs surface temperature (on which
ground structures which are temperature abundant meteorological data is available). 5 – USE OF ROCK MASS
critical during their service life (mined sto- Other very localised sources of heat may CHARACTERISATION FOR
rage) or need special cooling systems to be be chemical reactions, radioactivity, hot UNDERGROUND STRUCTURE
provided during and after construction water from deep underground, or cold STABILITY ANALYSIS AND
(deep tunnels and storage facilities). water from the surface. CONSTRUCTION
The estimate is frequently based on a
4.5.2 – Methods for estimating
mean value of the local geothermal gra-
temperatures for underground 5.1 – CHARACTERISTIC
dient Γ, which usually ranges from 2° to 4° VALUES AND PARAMETERS
structures
per hundred metres depth, although it may FOR DESIGN
take other values in tectonic zones and vol- At the project planning stage, the designer
canic ground. is usually confined to a desk search for
regional geothermal data (geothermal flux 5.1.1 – Individualisation of sub-
4.5.1 – Geothermal parameters map of France, Mechler 1982; isogradient units
maps, Gable 1980) and temperature data
The most important thermal property of a As mentioned in para. 1.2.2, the first step is
from oil wells and geothermal boreholes.
rock mass is its conductivity λ (in W/m.°C). to divide the rock mass up into sub-units
Geothermal springs should be identified
Values for this parameter for samples of dif- where they exist, along with nearby volca- exhibiting substantial uniformity in their
ferent rocks are confined to the range 1.5 nic ground (“hot” anomalies), and the pos- characterisation parameters. A sub-unit
to 5.5 W/m.°C (cf. Appendix 12). Because sibility of there being aquifers near the may correspond to a section of the tunnel
of the low porosity of rock, the presence of underground structure site supplied from (for linear structures) or part of the under-
water and air has very little impact on the surface water (“cold” anomalies). ground structure (for non-linear structures).

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 25

 Caracterisation of rock masses useful for the design and the construction of underground structures

Individualisation of a sub-unit generally at least of the order of 10. This analysis, • Swelling parameters σg and Cg, for which
starts with the lithology and geological with number of data points, maximum and the pair of characteristic values must lie at
structure as seen on the tentative geologi- minimum values, mean and standard the upper end of the measured values,
cal cross section; however, the lithology deviation, makes it possible to assess scat- • Uniaxial compressive strength σc, tensile
may admit of sub-divisions by reason of ter and the possible existence of several strength stb and Franklin index Is, for which
variations in homogeneous populations, as may hap- the characteristic value may be
• the mechanical properties of the material pen with an anisotropic rock or a material
consisting of thin alternating layers of diffe- - either the 5% fractile when focusing on
(with the same petrographic signature), the stability of the structure,
rent lithologies, e.g. marl-limestone.
• alteration and weathering, - or the 95% fractile when dealing with
For each uniform sub-unit, these manipula-
• hydrogeology, tions can be used to determine a 'characte- drillability.
• rock cover, ristic' value for each parameter. The charac- When dealing with stability, caution dic-
teristic value of a parameter represents a tates giving consideration to the case
• joint density, etc. reasonably cautious value, not a maximum where rock strength might locally be lower,
In practice, this operation of dividing the or minimum (cf. AFTES GT 29 whereas with drillability, attention focuses
underground structure up into uniform Recommendations on the use of general on the higher values of σc in order that tun-
sub-units usually involves several iterations design standards and rules for reinforced nel driving aspects are not underestima-
before the resulting breakdown emerges and plain concrete tunnel linings). ted. In all situations, a reasonably prudent
as entirely logical and coherent. As it is not For any one parameter, it may be useful to choice of characteristic value must be the
possible to set boundaries between classes set one or more characteristic values accor- outcome of a thinking process that must be
of values for the different parameters used ding to the intended use. For example, a explicit and substantiated on the basis for
as the most pertinent criteria for individua- low characteristic value for the uniaxial example of a statistical analysis, site speci-
lisation, a trial and error process is gene- compressive strength would be needed for fics, reference to completed structures in
rally necessary. similar settings, etc.
the stability analysis and a higher value for
the drillability assessment. • Cohesion C and friction angle φ, for which
5.1.2 – Geotechnical characterisation the analysis preceding determination of
of sub-units 5.1.2.1 – Rock matrix the pair of characteristic values C and φ
Consider the measurements and tests for a The description for non quantitative para- must include for all the circles determined
given sub-unit. meters (common name, petrography and on the same material, with no distinction
mineralogy, alteration of the rock matrix between one test and another.
Once site investigations have been made,
the raw data consists of measured values material) must be precise and, where
5.1.2.2 – Discontinuities
from tests performed by the geotechnical necessary, supplemented with quantitative
operator. values (mineral contents, methylene blue Before determining characteristic values
test value, etc.). qualifying discontinuities, the designer
Depending on the geotechnical terms of must address the following points:
reference (cf. French standard P 94-500), For most physical identification parameters
the next step is for the geotechnical opera- (density ρ, ρd, ρs or volume weight γ, γd, γs, • Joint density may itself be an individuali-
tor and/or designer to proceed with a criti- moisture content w, porosity n, wave velo- sation criterion for the sub-unit, and one
cal analysis of these measured values. This city Vp and continuity index IC) and for must always ask whether it is better to take
critical analysis eliminates any anomalous hardness DU and abrasiveness AIN or ABR, a mean value or consider two sub-units ins-
data and validates the significant values to the arithmetic mean of the values obtained tead of only one.
be retained. is the starting choice for the characteristic • Values found for the overall indexes
value, provided that the statistical analysis (RQD, ID, FD) may be closely dependent
The significant values relate to a type of confirms that the population is normal. on the orientation of the survey line.
test and a in-situ test:
The choice is not valid in the following When determining a characteristic value to
• They may be 'local' values associated cases: quantify discontinuities, the arithmetic
with a type of test or measurement at a
• Discovery of more than one population of mean of overall index values (RQD, ID, FD)
given sampling or in-situ test;
values (marl limestone, schists, etc.) measured in the same direction is satisfac-
• They may on the contrary be 'global' tory provided scatter is reasonable. The
values, associated with a type of test or • Variation linked to the spatial distribution most appropriate survey line direction
measurement, on a wider geographic area of samples must be chosen with reference to tunnel
(such as geophysical test data for • Severe scatter militating for caution in alignment and this is the direction along
example). favour of a lower characteristic value than which the characteristic value is determi-
It may be useful to make a sort of 'consoli- the mean minus standard deviation. ned.
dation' of the reliability of the significant With parameters for mechanical behaviour, For parameters describing discontinuities
values by using simple but proven correla- a distinction must be made between the in the same joint set, a distinction may be
tions such as E/Vp, σc/Vp, σc/γd, σc/σtb, etc. following: made between
to detect any anomalies. • Young's modulus E and Poisson's ratio n • strike OR, whose characteristic value can
Lastly, a quick statistical analysis should be which admit of the same approach as for be taken as the most frequently occurring
run if the number of data points available is the physical identification parameters, data point on the Schmidt density chart,

26 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

• joint spacing ES, whose characteristic • analysis of displacements measured in Terzaghi 1946, Lauffer 1958, Deere 1964,
value can be taken as the arithmetic mean the exploratory adit. Wicham 1972, Bieniawski 1973, and
in simple cases displaying little scatter; in Determination of the characteristic value Barton, Lien & Lude 1974. The Bieniawski
other cases, the consequences of variable then becomes a "reasoned" choice among and Barton systems are by far the most
spacings must be examined in detail, different significant values from the various widely used.
• joint persistence, whose value can be tests and measurements. The choice must Geotechnical classification systems are
approximated by observations on outcrops however take into account based on an empirical rock mass "quality
but whose characteristic value can only be • scale effect, illustrated by variations in the score" drawn from values determined for
the result of engineering judgement and field modulus with the rock volume consi- certain design-critical parameters. The
experience, dered, and by the empirical rule parameters involved vary slightly from one
system to another but are basically
• joint wall roughness, waviness and wea- EL > ED > EP > EG
thering, joint width, infilling and water if • rock matrix strength
in which EL is the laboratory modulus, ED is
any, for which, once again, engineering • joint density
the dilatometer value, EP is the plate loa-
judgement only can arrive at an average ding test value, and EG is derived from adit
characterisation for the whole joint set. • mechanical behaviour of discontinuities
wall displacements;
• hydrogeological conditions
5.1.2.3 – Rock mass • strain range: compared to other modulus
values and means of measuring them, the • state of stress (partially).
The rock mass may be characterised either
dynamic modulus is based on very small The scoring process produces a final value
directly from the results of appropriate in
strains. obtained through a simple calculation, which
situ tests, or indirectly with the aid of empi-
A mean permeability value can only be esti- also differs from one system to another.
rical classifications and correlations, relying
mainly on the characteristic values determi- mated in so far as local values exhibit little In 1978, at the time of first writing these
ned from laboratory samples of the intact scatter, to ensure there is a good probability Recommendations on the description of
rock material as well as all other sources of of their belonging to the same sub-unit. If rock masses, AFTES adopted a restrictive
data (geophysics, borehole tests and mea- this is the case, the mean is calculated on position towards these systems, arguing
surements, etc.). the log K values (log normal distribution). If that quantifying rock mass quality by
there is significant scatter in the data or means of a single score was too reductio-
Because of the time and cost involved, in significant differences between two or more nist and did not reflect the complexity of
situ tests are usually kept for a relatively test results, the designer should ask whe- the real world.
late stage in the design process once the ther it would not be more appropriate to
project layout has been more or less finali- consider several units with their own speci- Now these systems are widely used to
sed. In the earlier stages, indirect methods fic hydrogeological characters. derive, via the various correlations propo-
as discussed in para. 5.2 – Geotechnical sed, mechanical parameters for rock
By the same logic, where several in situ masses (modulus, Hoek & Brown coeffi-
Classifications and 5.3 - Correlations are
stress measurements are available, it might cients, etc.) which can be used as design
mostly used.
be preferable to select the stress state that input (see para. 5.3.1). It is nevertheless
in situ tests proper in boreholes, shafts and agrees best with drilling records (disking,
extremely important to remain sceptical
adits aim primarily at determining rock mass ovalisation, wall failure).
about the simplifying assumptions inherent
deformability, in situ state of stress and Generally speaking, the approach to obtai- in these systems and the choice of data on
hydrogeological conditions. Knowledge of ning characteristic values is to find and ana- which they are based.
the rock mass will be more or less extensive lyse the greatest number of cross-checks
and precise, depending on the resources Using a classification system for any parti-
between data from different sources in
assigned and the size of the project. There cular project presupposes that the desi-
order to arrive at a considered judgement
can be no doubt that an exploratory adit as to the correct characteristic value. gner has first assured himself that the pro-
driven over part or all of the alignment of ject is truly compatible with the system
the permanent structure will yield more, In the absence of any direct means of used (cf. para. 5.2.4).
determination, when significant values
and more precise, data than many bore- Furthermore, a classification system must
have not yet become available (as is often
holes and shafts. The designer will also never be considered as a substitute for site
the case with rock mass deformability and
have at his disposal several different ways investigations or be an excuse for cutting
nearly always with rock mass limit strength),
of arriving directly or indirectly at any given down on efforts to arrive at the geotechni-
a possible approach is to refer to a similar,
parameter and comparing the various cal characterisation of the rock mass. None
completed, structure and/or rely on classifi-
values obtained, assessing scatter and cation systems, always provided that the of these classification systems are univer-
scale effect. project falls within their category of validity. sally applicable.
For example, the rock mass deformation
modulus can be approached in various ways: 5.2 – GEOTECHNICAL CLASSIFICA- 5.2.2 – Bieniawski's Rock Mass
TIONS Rating
• borehole dilatometer tests
The Rock Mass Rating (RMR) has been
• plate loading tests on adit or shaft walls 5.2.1 – General developed by Bieniawski since 1973 to
• dynamic moduli derived from wave velo- Various authors have proposed classifica- provide a quantitative estimate of the pro-
cities obtained by seismic methods tion systems such as Protodiakonov 1909, perties of the rock mass and support

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 27

 Caracterisation of rock masses useful for the design and the construction of underground structures

necessary for stability. This approach was would automatically exclude rock strength The Barton Q index is written
initially based on records from more than classes RC6 to RC7 and stress class CN3. Q = (RQD/Jn) x (Jr/Ja) x (Jw/SRF)
300 tunnels, most of them lying at mode- The rating system produces five rock mass
rate depth in sedimentary rock. The data- In other words, the Q index is the product
classes (Appendix 13) and five correspon- of three factors for
base was based primarily on South African ding support classes, and this is nowadays
experience but has since been considera- inadequate to cover the variety and pro- • the potential size of rock blocks
bly enriched from many examples round gress encountered in excavation and sup- • the geomechanical quality of the contact
the world. After the first version had been port techniques. surfaces between blocks
widely circulated in 1976, Bieniawski made • the initial state of the rock mass as regards
many changes to the parameters for esti- 5.2.3 – Barton's Q index water and stresses (Barton's "active stress").
mating RMR. The current version, descri-
The Q index is the central parameter in a Calculating the range of variation of Q, first
bed here, is the RMR89 (Bieniawski 1989).
system developed by the Norwegian with the most unfavourable values, then
The RMR index is the sum of five scores Technical Institute in 1974 based on data with the most favourable values, may pro-
quantifying five characteristic rock mass from more than 200 completed tunnels, duce very large differences if the calcula-
parameters and an adjustment factor mostly situated in the crystalline tions are done for sub-units displaying very
dependent on azimuth and dip of the dis- Scandinavian Shield with high horizontal different characteristics.
continuities. The RMR has been calculated stresses (Barton et al. 1974). The system
to span the range 0-100. Table 23 recapitulates the ranges of varia-
was revised in 1993 to include data from
tion of the different parameters to assess
The five ratings A1 to A5 and rating adjust- more than 1000 tunnel case histories
their relative weight in the final Q index
ment B (cf. Appendix 13) are defined as (Grimstad & Barton 1993).
value.
follows: The Q system method provides a quantita-
The weight of the SRF factor in the third
- A1: Strength of rock matrix: Range of tive estimate of support needed for tunnel
term Jw/SRF is particularly high, which is
values 0 to 15 based on uniaxial compres- stability on the basis of the following infor-
the unique feature of the Q index, which
sive strength or point load strength Is. mation:
refers to :
- A2: Drill core quality: Range of values 3 to • Largest dimension (diameter) of the plan- • the possibility of sheared, brecciated or
20 for rock core quality, from RQD. ned opening very clayey zones;
- A3: Spacing of discontinuities: Range of • Planned use of the completed structure • the level of stress in brittle rocks;
values 5 to 20 (lowest ratings for each joint (implicitly, acceptable level of risk)
• potential creep and swelling stresses in
set). • Rock mass Q index. deformable rocks.
- A4: Condition of discontinuities: Range of The Q index is a total score from 0.001 to The Q index is thus strongly dependent on
values 0 to 30 (joint persistence, width 1000 (this is the theoretical range, reduced non-intrinsic rock properties, especially the
(separation), roughness, infill (gouge) and in most practical cases to 0.005-50), calcu- state of stress in the rock mass. The formu-
wall rock weathering). lated from (Appendix 14) lation of the Q index does however have
- A5: Groundwater: Range of values 0 to 15 • Rock Quality Designation (Deere 1964) the drawback of not directly reflecting the
(inflow rate and/or pressure). characteristic parameter of the mechanical
• Joint set number Jn strength of the rock material.
- B: Adjustment for joint orientation: Range • Joint roughness number Jr (concerns the
of values –12 to 0, for strike and dip of dis- most unfavourable discontinuities) 5.2.4 – Summary and precautions
continuities with respect to tunnel align-
• Joint alteration number Ja (concerns the The growing popularity of classification
ment.
most weathered discontinuities and infill systems in France is probably due to :
The basic Rock Mass Rating (RMRbasic) cha- material) • their apparent simplicity of use;
racterising the rock mass is simply the sum
• Joint water reduction factor Jw (flow rate • their very widespread use outside France,
of ratings A1 to A5 (B = 0). In underground
and pressure) especially by French engineers working
engineering work, the standard RMR (or
RMR89) is written as • Stress reduction factor SRF. abroad;

RMR89 = A1 + A2 + A3 + A4+ A5 + B
Basically therefore, RMR is a rating assi- PARAMETERS MOST UNFAVOURABLE MOST FAVOURABLE RANGE
CONDITIONS CONDITIONS (highest ratio)
gned to the rock mass ranging from 0 to
100, more than 70% depending on discon- RQD 10 100 10
tinuities and only 15% on rock matrix pro- Jn 20 0,5 40
perties and 15% on hydrogeology. The Jr 0,5 4 8
rating completely ignores the state of Ja 20 0,75 27
stress in the rock mass at the tunnel site.
Jw 0,05 1 20
This should theoretically limit the use of the SRF 20 (6) 0,5 40
RMR only to strong rock whose response is
governed by the discontinuities. This Table 23 – Ranges of variation of parameters used in calculating the Barton Q index

28 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

6 usual classification systems. Correlations "in

cascade" must never be used.
PARAMETERS RMR Q system

5.3.1 – General
• Jointing patterns well descri- • Mechanical properties of dis-
Overall characterisa- bed except for anisotropic rock continuities well described In view of the difficulties of making direct
tion of rock mass (schist, slate, etc.) tests of deformability and (even more so)
• Natural stresses described
limit strength at rock mass scale, many
authors have sought to start from actual
Assessment of • Empirical correlations bet- • Empirical correlations between
case histories to establish empirical rela-
mechanical characte- ween RMR and deformability Q and physical and mechanical
and strength parameters tionships linking these parameters to rock
ristics at scale of parameters (P longitudinal wave
matrix characteristics and rock mass join-
whole rock mass velocities, deformability)
ting.
• Must be used with great caution, especially for strength
parameters: avoid correlations 'in cascade' of the type These relationships have been established
Q ⇒ RMR ⇒ (m, s) ⇒ (C, ϕ) for particular contexts and must be used
with great caution; they must always, as far
• Allowance for orientation of • Not relevant to orientation of
as possible, by set side-by-side with in situ
discontinuities with respect to discontinuities with respect to
field test results.
axis structure centreline
• Quick means of setting length • Quick means of stipulating 5.3.2 – Estimating rock mass
Use for project of pull support needed at roof, side- deformability
• Stand-up time (conservative walls and intersections but gives The rock mass deformation modulus EMas is
approach) false impression of accuracy in
one of the critical parameters for modelling
setting bolt lengths
• No use in deciding excavation stresses and strains around an under-
method • Use in design stage and for ground opening. There are several means
monitoring tunnel driving of measuring this parameter at a volume
• Allows for changes in support scale of up to a few cubic metres (see para.
techniques 4.2.1.2) and estimating it at a larger scale
(but for very small load values) by seismic
Table 24 – Comparison between RMR and Q system in underground engineering applications tests (para. 4.2.1.1). As already stated in
para. 5.1.2.3, the scale effect is very impor-
tant here.
• the convenience of using a rating system systems must be used with the following
Many schemes have been proposed for
for making comparisons between design precautions:
indirectly estimating the rock mass defor-
predictions and actual conditions encoun- • Do not use only one system. mation modulus. The more important ones
tered during construction at different sites;
• Explain in detail how the scores were cal- are tabulated in Appendix 15, along with
• the possibility of amending scores in the culated; most importantly, identify the joint author references.
light of conditions encountered during sets considered at each step. These schemes may directly combine para-
construction;
• Examine the sensitivity of the RMR or Q meters for the rock matrix (E, σc) and rock
• the possibility of using correlations to index to changes in the values of the para- mass (RQD) or be derived indirectly via the
find the quantitative data need for design meters and present results as envelope RMR and/or Q index.
analyses. values for the final rating. Figure 14 shows a few examples of the
In underground engineering, the ultimate • Do not use the ratings as a "rule-of- empirical relationship between E moduli
purpose of these classification systems is thumb recipe," but be critical and vigilant and RMR and Q index.
to design tunnel support; this approach as to the proper field of application.
has been tried and found satisfactory in 5.3.3 – Hoek's GSI index
much drill and blast tunnelling. But these • Remember that classification systems are
empirical and reflect certain tunnelling and The Geological Strength Index is not
systems are not always suitable with other
support practices, and these practices may directly a classification system, it is an inter-
excavation methods (road headers, tunnel
change. mediate step to determining the mechani-
boring machines).
cal properties of a rock mass, using the
Generally speaking, the RMR and Q sys- empirical formulae proposed by Hoek &
tems are unsuitable for soft rock (R6 to R7). 5.3 – CORRELATIONS
Brown (see below).
Table 24 summarises the features and limi-
Warning: It must never be forgotten that trea- GSI was introduced in 1995 by Hoek
tations of these two systems.
ting a jointed rock mass as a continuum mate- (Strength of Rock and Rock Masses, ISRM
In addition to the general and specific limi- rial is in itself a considerable simplification. News Journal, 1994, vol. 2). It derives from
tations discussed above and in Table 24, it Secondly, highly anisotropic rock masses dis- variants of the RMR and/or Q index, desi-
must also be stressed that classification play special behaviour not covered by the gnated RMR' and Q' respectively.

6
Much higher values, up to 400, have been suggested by Barton for very deep underground openings where there is a risk of sudden violent decompression

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 29

 Caracterisation of rock masses useful for the design and the construction of underground structures

low-friction rock masses to 0.4 to 0.6 for

hard rock containing few, very rough-wal-
led joints.
There are also formulae proposed by
various authors (Hoek 1990) to derive fric-
tion angle and cohesion values from coeffi-
cients s and mb. They are obtained by
linearising the parabolic criterion and
replacing it by a tangent or secant over a
set stress range. Despite the attractions
this may appear to offer, making it possible
to extend standard soil mechanics calcula-
tions based on the linear Mohr-Coulomb
criterion to rock masses7 , it must be stres-
sed that this criterion is generally irrelevant
to the characterisation of rock mass beha-
viour. It may produce quite suspect results
Figure 14 – Estimating rock mass deformation modulus EMas from RMR and Q values and must always be used with great cau-
(Hoek, Kaiser & Bawden 1997) tion and scepticism.
Lastly, this approach must never be used
when dealing with rock masses whose dis-
RMR' is calculated like the basic RMR but and Brown (brought together in Hoek, continuities are strongly polarised (thinly
using only the first four criteria (Strength, Kaiser & Bawden 1977) who suggest bedded rock, schist, slate, etc.): assigning
RQD, Spacing of Discontinuities and extending the parabolic failure criterion an isotropic failure criterion derived from
Condition of Discontinuities), systemati- proposed for the rock matrix (para. 2.2.5.2) the RMR to a situation where actual beha-
cally taking the fifth groundwater value as to the rock mass, suitably modified to the viour is strongly anisotropic can only lead
15 (it is rock behaviour under "completely following generalised form: to results that have little relationship with
dry" conditions that is considered) and the σ1 = σ3 + σci [mb . σ3/σci + s]a (1) reality.
rating adjustment for joint orientation as 0.
in which a, s and m b are characteristic
By a similar process, Q' ignores the third constants for the rock mass.
ratio for the fifth and sixth parameters
(water and 'active stress'). Except in highly weathered rock with prac-
tically no remaining cohesion, the value
Q' = (RQD/Jn) x (Jr/Ja) generally adopted for a is
GSI is determined from RMR'89 values as a = 1/2 (parabolic criterion).
follows:
Values for the constants can be derived
• for RMR'89 > 23 from equations containing Hoek's GSI (see
GSI = RMR'89 – 5 para. 5.3.3).
• for RMR'89 < 23 mb = mie[ ( GSI x 100)/28]
GSI = 9(Log Q' + 44). For GSI > 25
s = e[(GSI x 100)/9]
5.3.4 – Estimating rock mass limit
strength a = 0.5
Estimating rock mass limit strength at For GSI < 25
underground structurescale calls for fine s=0
judgement. As mentioned above (para.
4.2.2) no in situ test – except a full size test a = 0.65 x (GSI/200)
to failure, which is unfeasible for obvious The relationship between mb and mi, the
reasons – is capable of yielding useable rock matrix characteristic constant close to
results. The only possible approach is to the brittleness index FR (see para. 2.2.5.2)
Anisotropic heterogeneous rock mass –
downscale, on empirical evidence, the pro- is very important for calculating numerical Mequinenza region, Spain
perties of the intact rock matrix with refe- values in equation (1). From Hoek &
rence to rock mass jointing. The most signi- Brown's compilation, the ratio mb/mi may Photo 9 – Anisotropic heterogeneous rock mass –
ficant research in this area is due to Hoek range from low values (<0.1) for jointed Mequinenza region, Spain

7
The success that this has had with many sometimes unsuspecting users is due to the ease it offers them when having deal with rock mass problems. But
such popularity cannot in any way be advanced as an argument for the scientific merits of the approach.

30 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

5.4 – PRESENTATION OF ROCK

MASS CHARACTERISATION
DATA

5.4.1 – Basics and general

remarks
The variety of geologies and geotechnical
conditions which may be encountered in Figure 15
underground engineering and the specific Principle of segmentation
features of each project make it difficult to into homogeneous geo-
technical sub-units and
envisage any single format for presenting data hierarchy
synthetic data suitable for all possible
cases. The general presentational scheme
suggested here will have to be amended
as required for each individual project.
Data presentation may differ at different
stages of project development (project
planning, preliminary design, final design,
etc.) and according to the quantity and
reliability of the data available.
When summarising data characterising the
rock mass, one must bear in mind the ulti- 'basic minimum' to be collected for any basic document bridging the gap between
mate destination of the data (general underground project. Other parameters geologists and geotechnical engineers,
design, construction method, design ana- may be added as required for any specific and civil engineers. The most common
lyses, etc.) and the fact that some (or all) project (e.g. swelling pressure σg and swel- method is to have a single drawing with the
items may have contractual relevance, ling index Cg for potentially swelling geological profile at the top and the break-
depending on the stage reached in the rocks, etc.). down into sub-units below, along with the
project implementation process: the form For each homogeneous sub-unit, the sum- summary table. The geological profile
and content of the summary presentation mary table may include the AFTES classes should be drawn to true scale so as not to
must consider this point. for rock matrix, discontinuities and rock give an inaccurate picture of the structures
mass parameters. One should not however with distorted contacts and dips.
5.4.2 – Example of data presenta- be tempted to have so many items as to
tion in tabular form 5.4.3.2 – Cross sections and horizontal
make what is a 'summary' table difficult to
In order to remain consistent with the hie- read, because the point of a summary is to section at project depth
rarchy of the classes for the various parame- be clear and easily understood. The complexity of the geological and geo-
ters examined in these Recommendations, technical context may make it extremely
NOTE. Figures shown in the E4 column of
the homogeneous sub-units determined useful or essential to accompany the geo-
Table 25 and details in the Orientation line
from the field investigations and studies logical longitudinal profile in some places
for discontinuities are not taken from an
should be ordered into geotechnical with cross sections. This is particularly the
actual project and are shown simply for
classes from E1, representing the globally case when structural details of significant
illustrative purposes.
highest geotechnical classes, to EN, repre- importance for the project are strongly ske-
senting the poorest geotechnical proper-
5.4.3 – Synoptic presentation of wed in relation to the tunnel. Again, under
ties (see figure 15). It must be stated whe- rock mass characterisation data some conditions when the geological lon-
ther the values for these parameters are and cross-referencing to geological gitudinal profile alone is unable to give a
significant values or characteristic values (cf. profile true picture, a horizontal geological profile
para. 5.1.2 above).
at project depth should be added for a
Naturally, this is a recommendation which 5.4.3.1 – Longitudinal profile better understanding of the geological
one should try to follow, sometimes with a As an aid for using the parameters charac- context and potential risks: this situation
little difficulty in that certain parameters terising the homogeneous geotechnical arises for example when the tunnel align-
examined may act in contrary directions, sub-units for the various uses to be made ment is expected to draw near to, without
making it more problematical to arrive at a of the data (project design, construction necessarily intercepting, a major structural
perfect hierarchy. methods, etc.), the summary table is features (fault, unconformable contact,
The parameters listed in the example sum- usually shown alongside the geological permeability interface, etc.) running near
mary Table 25 must be considered as the profile which, for any given project, is the and parallel to the tunnel.

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 31

 Caracterisation of rock masses useful for the design and the construction of underground structures

Tableau 25 :Tableau de synthèse des paramètres de caractérisation des sous-ensembles géotechniques homogènes

32 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX

CONTENTS
Pages Pages

APPENDIX 1 APPENDIX 10
RECOMMENDED ROCK DESIGNATIONS AND HYDRAULIC PARAMETERS - - - - - - - - - - - - - - - - - - - - - - - - 41
PRINCIPAL GROUPS - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 34 APPENDIX 11
APPENDIX 2 RIGID PLATE LOADING - - - - - - - - - - - - - - - - - - - - - - - - - - - 41
DENSITY AND THEORICAL P WAVE VELOCITIES VP APPENDIX 12
IN MINERALS - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 34 THERMAL ROCK PARAMETERS- - - - - - - - - - - - - - - - - - - - - 42
APPENDIX 3 APPENDIX 13
ORDER OF MAGNITUDE THEORETICAL P WAVE BIENIAWSKI’S RMR CLASSIFICATION OF ROCK MASSES - 43
VELOCITIES VP* INSOME ROCKS ASSUMED TO BE
APPENDIX 14
SOUND AND NON POROUS- - - - - - - - - - - - - - - - - - - - - - - 35
BARTON’S Q INDEX - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 45
APPENDIX 4
APPENDIX 15
SWELLING TESTS - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 35
empirical formulae for evaluating rock mass
4.1 – Huder-Amberg test - - - - - - - - - - - - - - - - - - - - - - - - - - 35
deformation moduli - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 47
4.1.1 – Test procedure - - - - - - - - - - - - - - - - - - - - - - - - - - 35
4.1.2 – Interpretation- - - - - - - - - - - - - - - - - - - - - - - - - - - 36 REFERENCES - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 47
4.2 – ISRM tests - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 36 16.1 – General - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 47
4.2.1 – Axial swelling pressure at constant volume - - - - - 36 16.2 – Rock matrix characterisation parameters- - - - - - - - - - 47
4.2.2 – Axial swelling pressure versus axial strain - - - - - - 36 16.3 –Characterisation parameters of discontinuities - - - - - - 47
16.4 –Set persistence of discontinuities and 3D
APPENDIX 5
geometrical modelling of networks of discontinuities
MEANING OF PARAMETERS IN HOEK & BROWN - - - - - - - 37
(Appendix 8) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 48
APPENDIX 6 16.5 – Schmidt hammer test (Appendix 9)- - - - - - - - - - - - - - 48
CHART DEFINING DRILLING RATE INDEX AND 16.6 – Hydraulic parameters (Appendix 10) - - - - - - - - - - - - - 48
CHARACTERISTICS OF S 20 and sj tests- - - - - - - - - - - - - - - 37 16.7 – Rock mass characterisation parameters- - - - - - - - - - - 48
6.1 – Laboratory tests- - - - - - - - - - - - - - - - - - - - - - - - - - - - - 37 16.7.1 – Indirect measurements - - - - - - - - - - - - - - - - - - - 48
6.2 – Interpretation - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 38 16.7.2 – Rigid plate loading test- - - - - - - - - - - - - - - - - - - 48
APPENDIX 7 16.7.3 – Stresses - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 48
TEREOGRAPHIC PROJECTION - - - - - - - - - - - - - - - - - - - - - 38 16.7.4 – Thermal rock parameters (appendix 12) - - - - - - 49
APPENDIX 8 16.7.5 – Hydrogeological conditions - - - - - - - - - - - - - - - 49
set persistence and 3 D GEOMETRICAL MODELING OF 16.8 – Use of rock mass characterisation for design and
FRACTURE NETWORKS - - - - - - - - - - - - - - - - - - - - - - - - - - 39 construction of underground structures - - - - - - - - - - - - - - - 49
16.9 – AFNOR standards - - - - - - - - - - - - - - - - - - - - - - - - - - 49
APPENDIX 9
SCHMIDT HAMMER - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 40

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 33

APPENDIX Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX 1

RECOMMENDED ROCK DESIGNATIONS AND PRINCIPAL GROUPS

Granite, granodiorite, syenite, microgranite, rhyolite, rhyodacite, tra-

Granites
chyte, etc.

IGNEOUS ROCKS Diorites Diorite, quartzitic diorite, microdiorite, andesite, dacite, trachyandesite,
etc.
Basalts & Gabbros Gabbro, dolerite, peridotite, serpentine, basalt, pozzolana, etc.

Massive metamorphic rocks Gneiss, amphibolite, cornelian, quatzite, marble, leptynite, etc
METAMORPHIC ROCKS
Schistose metamorphic rocks Schist, micaschist, slate, calcschist, etc.
Carbonate rocks Limestone, chalk, dolomite, cargneule, travertine, marl, etc.
SEDIMENTARY ROCKS
Detrital rocks Sandstone, arkose, claystone, pelite, conglomerate, etc.
Saline rocks Rock salt, gypsum, anhydrite, potash, etc
Carbonaceous rocks Coal, lignite, etc.
Names in italics are extrusive or volcanic equivalents

APPENDIX 2

MASSE VOLUMIQUE ET VITESSE THEORIQUE VP DES ONDES P

DANS LES MINERAUX

Minerals Density ρs (g/cm3) Vp (m/s)*

Amphiboles 2.98 - 3.20 7 200

Augite 3.2 - 3.4 7 200
Biotite 2.90 5 130
Calcite 2.71 6 660
Dolomite 2.87 7 900
Magnetite 5.17 - 5.18 7 410
Muscovite 2.83 5 810
Oligoclase 2.64 - 2.67 6 260
Olivine 3.25 - 3.40 8 400
Orthose 2.57 5 690
Quartz 2.65 6 050
*from French standard P 18-556

34 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 APPENDIX
Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX 3

ORDER OF MAGNITUDE THEORETICAL P WAVE VELOCITIES VP* INSOME

ROCKS ASSUMED TO BE SOUND AND NON POROUS

Rock Vp* (m/s)

Granites and rhyolites 6 000
Diorites 6 500
Gneiss 6 000
Amphibolites 6 500
Calcaires 6 500
Silica rocks 6 000
* A utiliser au besoin dans la détermination de l’indice de continuité Ic

APPENDIX 4

SWELLING TESTS

4.1 – Huder-Amberg test Wetting causes the material to expand, D representing the state of the sample
causing by an increase in sample height ∆h after stabilisation of volume expansion.
under stress σD. The change in height ∆hD is
4.1.1 – Test procedure C – Third step: Decremental unloading
recorded over time until there is no further
A – First step: Reconfinement of sample change and ∆ h D remains From point D, the load on the sample is
constant. instantaneously reduced from pressure σD
The sample of height h is accurately dres-
On figure 1, this change to σE; the immediate state of the sample
sed to fit snugly in the test cell. It is placed
in sample thickness at can now be represented by point E’. The
in the oedometer between porous stones.
constant pressure effect of load reduction is to allow swelling
The top stone is in contact with the piston
applying a pressure opposing all (or part σD appears as the to proceed in the form of a further volume
straight seg- expansion causing a change in sample
of) the increase in sample height ∆h. The
ment D’-D, height ∆hE. As before, the change is recor-
test starts with only the weight of the pis-
with point ded to stabilisation, at which point, the
ton exerting a very low stress σm of the
state of the sample can be represented by
order of 0.025 MPa, considered as the ori-
point E, the change in height of the sample
gin on the semi-log paper on which the
due to unloading from stress σD to stress σE
test is plotted.
is represented by straight line E’ – E.
Any imperfections (decompression, micro-
This is repeated several times, reducing the
cracking) from the sampling or sample pre-
stress s each time and, each time, waiting
paration process are corrected for by:
for stabilisation of the height increase ∆h.
• applying load (a) to produce a stress σD
Note. For consistency with ISRM
equivalent to the in situ stress;
recommendations (see para. 1.4), the
• then decreasing this load (b) to stress σm; unloading should follow a geome-
• increasing the load (c) to σD (if the sample trical function:
was unaffected by sampling, the points for σi+1 = 0.5 σi = 0.25 σi-1
the two σD load conditions coincide).
and limit load decre-
B - Second step:Wetting ments to a recommen-
ded minimum value of
The sample is them brought into contact 25 kPa.
with water through the bottom porous
stone, with the top porous stone allowing Appendix 4 – Figure 1 – Huder-Amberg swelling test
air in the sample to escape.

TUNNELS ET OUVRAGES SOUTERRAINS -TUNNELS

N° 177 - ET
MAI/JUIN 2003
OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 35
APPENDIX Caracterisation of rock masses useful for the design and the construction of underground structures

4.1.2 – Interpretation ∆h/h = Cg x log (σj/σJ) The sample of height h is inserted in the
The test is interpreted in a semi-log plot in in which the swelling index Cg is a constant oedometer as in the Huder-Amberg test.
which the sample volume increase ∆h is governed by the intrinsic nature of the After wetting, the axial pressure on the
represented by the height change of the material. sample is controlled to oppose any height
sample in the oedometer expressed as a change ∆h in order to keep the sample
percentage of the initial height: B – Huder-Amberg meaning of ‘swelling volume constant. The test is continued
pressure’ until reaching the maximum pressure nee-
(∆h/h) x 100 ded to achieve this.
The swelling pressure as understood by
The changes in pressure on the sample are Huder & Amberg is defined as the stress σG
represented by the axial pressure on the beyond which wetting ceases to cause fur- 4.2.2 – Axial swelling pressure
piston, expressed in log σi. ther sample thickness change. This value is versus axial strain
given by the intersection of the Huder- The test procedure is exactly the same as
A – Huder-Amberg law
Amberg straight line S with the extension for the Huder-Amberg test except for the
Experience shows that, in this semi-log of the reloading curve c. way in which the test data is plotted.
plot, points D, E, F, etc. representing the
Note. The precision of this determination is Test results are plotted as a curve of per-
state of the sample after stabilisation of the
improved when stress σD on first wetting cent thickness change strictly due to swel-
height increase at each stage of load
(point D’) is very close to σG (this reduces the ling vs applied axial pressure.
reduction plot on a straight line. From this,
approximation of the extrapolation).
a very simple relationship can be found ‘Thickness change strictly due to swelling’
between the stress change ∆σ and sample means the total change resulting from the
4.2 – ISRM tests
thickness change ∆h. change in pressure minus pseudo-elastic
(see References, Appendix 16.2)
The thickness change ∆h when unloading strain corresponding to the same load
the sample from stress σi to stress σJ can be reduction.
4.2.1 – Axial swelling pressure at
written constant volume

APPENDIX 5

MEANING OF PARAMETERS IN HOEK & BROWN

Hoek & Brown (1980) wrote the intact rock From this, the relationship between the mi = FR
(rock matrix) parabolic failure criterion thus: Hoek & Brown coefficient mi, and σci and σti is This has the advantage of giving physical
σ1 = σ3 + σci (mi x σ3/σci + 1)1/2 σti = σci/2 x [mi – (mi2 + 4)1/2 ] meaning to coefficient mi and allows it to
and introduced constant mi, thereafter This is equivalent to be estimated from standard laboratory
commonly known as the Hoek & Brown tests. Uniaxial compression and Brazilian
FR1 = σci/σti = 2/[mi – (mi2 + 4)1/2 ]
coefficient. test results readily lend themselves to sta-
= 1/2 x [mi + (mi2 + 4)1/2 ] tistical analysis. The observation in para.
The standard form of the parabolic crite-
From the compilation of mi values (Hoek, 2.2 in the main text on anisotropy
rion giving components σn and τ of the
Kaiser & Bawden 1995), it is found that this obviously applies.
stress on the failure face versus σci and σti
coefficient ranges from 4 (for some clayey
(rock matrix compressive and tensile Finally, the parabolic criterion can be writ-
rocks) to more than 30 (igneous rocks and
strengths) is ten as
some metamorphic rocks). In practice the-
τ = σti [(1 + σci/σti)1/2 - 1] x (1 + σn/σti)1/2 refore, the following equation can be used: σ1 - σ3 = σci (FR x σ3/σci + 1)1/2

1
FR = brittleness index (see para. 2.2.4.3.

36 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 APPENDIX
Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX 6

CHART DEFINING DRILLING RATE INDEX AND CHARACTERISTICS

OF S 20 AND SJ TESTS

The Drilling Rate Index (DRI) was originally DRI is obtained from S20 and SJ using the
developed in the sixties at the Norwegian chart shown in figure 3. It ranges from 20
Institute of Technology at Trondheim to to 90. A high DRI value indicates easier
assess the drill hammer . Since the eighties, penetration of the TBM cutters.
it has mainly been used to predict the per-
formance of rock tunnelling boring
machines ( TBM )(see References,
Appendix 6 – Figure 3
Appendix 16.2). Chart for calculating Drilling Rate Index

6.1 – Laboratory tests

DRILLING RATE INDEX DRI

DRI is calculated from the results of two Example:

Norwegian standard laboratory tests:
• The S20 brittleness (fragmentation) test
(figure 1) Rock = gneiss of South Africa
This test estimates the resistance of the S20 = 45
rock to crushing under repeated blows, as
in the French dynamic fragmentation test.
SJ = 6.0
• The Siever J-value (SJ) penetration test
(figure 2)
This test estimates rock resistance to pene- From index A4.3, DRI = 43
tration, as in the Cerchar-Ineris hardness
test. More DRI value is ligh,
more the penetration of TBM is high
6.2 – Interpretation
BRITTLENESS VALUE S20

Appendix 6 – Figure 1 – Brittleness test Appendix 6 – Figure 2 – Siever test apparatus

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 37

APPENDIX Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX 7

TEREOGRAPHIC PROJECTION

The stereographic projection is the most of the reference sphere) or the projection observations represented on the radial
widely-used method of plotting disconti- of its trace (intersection of the plane with axis. This type of plot ignores dip angles
nuities. Various techniques are available the upper or lower hemisphere of the and is only meaningful for finding directio-
but two only are in routine use: sphere). nal sets of discontinuities with similar dip
• Equal angle projection to study relation- Two types of plots are routinely used to angles.
ships between discontinuities (Wülff net) analyse the pattern of discontinuities into • Density stereograms on the Schmidt net:
• Equal angle projection to measure spa- sets: densities are obtained by counting the
tial distributions (Schmidt net). • The cluster of azimuths of the dip vector: number of poles within the target 1% of
this consists of grouping observations in the diagram area. The count is done on a
Each joint plane is plotted as the projection counting net: count by circles centred on a
of its pole (intersection of the normal to the angular sectors of the dip vector azimuth,
with the absolute or relative number of regular grid or Dimitrijevic count by
plane with the upper or lower hemisphere ellipses centred on an irregular grid. The
trace of the joint number or density
contours bounds the zones of pole concen-
trations and may identify main joint sets.

Rose diagram of azimuth dip vector (example)

Représentation d’un plan (αp,β) à l’aide de Canevas de Wülff et de Schmidt Density stereogram on Schmidt net (lower hemisphere)
la projection stéréographique (hémisphère supérieur) with isovalues curves

Appendix 7 – Figure 1 – Cyclographic plot and polar plot of a plane of a discontinuty Appendix 7 – Figure 2 – Analysis of patterns
by stereographic projection of discontinuities (examples)

38 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 APPENDIX
Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX 8

SET PERSISTENCE AND 3 D GEOMETRICAL MODELING OF FRACTURE NETWORKS

The persistence of discontinuities can only have developed three-dimensional geo- write mathematical relationships between
be estimated from measurements on expo- metrical models of networks of discontinui- the 2D distribution of trace lengths and 3D
sed surfaces (outcrops, adit walls) and must ties. Among the more recent ones, the ran- distribution of disk radii (Warburton 1980).
be analysed with caution. dom disk model is one of the most widely
• The persistence distribution law: in this
Persistence is only accessible through two- used (Baecher 1977, Long 1985, Cacas
way, the parameters in this law can be
dimensional data – trace lengths – whose 1989, Xu 1992). For a given joint set, each
determined on the basis of trace length
measurement is affected by multiple geo- discontinuity is represented by a disk of
zero thickness, defined by the position of distribution parameters.
metrical bias factors. The first is that the
survey line preferentially intercepts the lon- its centre, its radius and orientation, each From this, one can obtain a 3D model of
ger discontinuities, the second is that the parameter being drawn stochastically from the network of discontinuities and study its
longer discontinuities extend beyond the its own distribution law, characteristic of connectivity to analyse the fluid flow
survey surface, introducing a truncation in the set. through the network or the mechanical
the measurements, the last is that some Estimating disk radii, i.e. the true extension behaviour of the assembly of rock blocks
limiting value is generally imposed on short of the discontinuities, raises problems. In created by the fractures, from fracture
discontinuities, which also has a truncating addition, passing from 2D continuity to 3D behaviour and, sometimes, the rock matrix
effect. Data must be corrected for any rigo- persistence is not simple and requires behaviour (figure 1).
rous estimate of this parameter (Priest & hypotheses on More complex models also introduce a
Hudson 1981, Pahl 1981). • The geometrical shape of the disconti- ranking of the joint sets (Heliot 1988) or use
Over the last thirty years, many authors nuities: in the disk model, it is possible to geostatics (Billaux 1990).

Appendix 8 – Figure 1 – Example of geometrical

modelling of network of discontinuities with
SIMBLOC program (ENSP-CGI, after Xu 1991)

MODELLED CONNECTED
DISCONTINUITIES DISCONTINUITIES

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 39

APPENDIX Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX 9

SCHMIDT HAMMER

The Schmidt hammer is an apparatus for mea-

suring the rebound as a mass strikes a surface,
powered by a compression spring, with known
energy. Rebound is measured as an index from
0 to 100.
Various types of hammer are available with dif-
ferent impact energies. In rock mechanics, the
most widely-used model is the Type L hammer,
for which correlations between the rebound
index and uniaxial compressive strength have
been established by Miller in 1965 (see
References, Appendix 16.5) (figure 1).
The measurement of the uniaxial compressive
strength of a joint wall is governed by a number
of procedural parameters (Barton & Choubey
1977):
• Hammer orientation: rebound is minimum
when the apparatus is held vertically down-
wards, and maximum when it is held vertically
upwards. Readings must be corrected when the
apparatus is used in other directions. In all
cases, it must be held perpendicular to the test
surface.
• Sample size: the sample must be large
enough for the impact not to cause it to move.
Small samples must always be fixed on a rigid
base.
• Number of measurements: experiences
shows that at least ten readings must be made
at different points on a representative sample or
per square metre area. The rebound value to be
used is the average of the five highest readings,
the lowest readings being considered as unre-
presentative because the samples are assumed
to have moved or the grains to have crushed.

Appendix 9 – Figure 1 – Relationship between Schmidt hammer rebound

and uniaxial compressive strength (after Deere & Miller)

40 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 APPENDIX
Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX 10

HYDRAULIC PARAMETERS

Fluid flow through discontinuities is a V is the mean fluid flow velocity in the dis- and depends on wall roughness. Various
highly complex problem. Experimental continuity semi-empirical formulae include for rough-
work (Gentier 1986) has shown it is not iso- Kf is the hydraulic conductivity of the dis- ness with the coefficient
tropic but follows channels whose geome- continuity ra/2a
try, while of course dependent on joint
width, is also governed by wall roughness J f is the orthogonal projection of the called relative hydraulic roughness, in
and their contact surfaces, by applied nor- hydraulic gradient on the fracture plane which ra represents the difference between
mal and tangential stress levels, and by the highest peak and the lowest trough in
a is the width or aperture of the disconti-
tangential movements of the walls, without joint wall geometry:
nuity.
forgetting the possible presence of filling Kf = (a2g/12ν) [1/{1 + B(ra/2a)1.5}]
With a smooth-walled planar discontinuity,
material. hydraulic conductivity K f(LT -1) depends with, for example, B = 8.8 (Louis 1969),
Detailed determination of the network of solely on its physical aperture and, by ana- B = 20.5 (de Quadros 1982).
channels and changes thereto due to shear logy with flow between flat plates, is written With ra ≤ 2a < 0.033, the effect of rough-
movements is therefore essential for Kf = (a2g)/12ν ness is negligible and the cubic law for
modelling the hydraulic behaviour of a dis- parallel plate flow can be used.
continuity, but this is as yet still at the in which
Barton (1985) established a correlation bet-
research stage. a is the physical aperture of the disconti-
ween relative hydraulic roughness and the
Various more or less simplified approaches nuity (L)
Joint Roughness Coefficient JRC. This led
are however available for estimating frac- g is the acceleration due to gravity (LT-2) him to introduce the concept of hydraulic
ture flow rate Q with the equation ν is the kinematic viscosity of the fluid (L2T-1) aperture A of a discontinuity, related to
Q = Va = Kf Jf a (for laminar flow) physical aperture a and JRC as
whence flow Q is proportional to the cube
in which of the aperture (cubic law). A = JRC2.5/(A/a)2
Q is the flow rate through the discontinuity In a natural discontinuity, the geometry of The dimensions of a and A are millimetres.
per unit width the voids between the walls is not constant

APPENDIX 11

RIGID PLATE LOADING

The test consists of applying load to the affected by the excavation process. A ball The plate must bear on a flat rock surface,
rock by jacking a rigid plate against it. The joint makes up for any misalignment and free from any loose fragments caused by
test is performed in an adit (2-2.5m section) lack of parallelism between the two faces. blasting, and dressed by bush hammering.
with the reaction being provided by the ∑ A two-speed pump and two 1% class Cement dressing should be kept to a mini-
opposite rock face, allowing two measure- pressure gauges. mum, never more than a few millimetres
ments to be made simultaneously. It is thick. If measurements are to be made with
usually performed either horizontally or ∑ A rigid reference frame fixed beyond the both plates, the bearing surfaces of the
vertically, although other set-ups may be zone of influence, carrying displacement rock must be strictly parallel.
justified in strongly anisotropic rock. gauges (C4, C5, etc.) to measure the dis-
placement of the plate and surrounding Places where the frame is anchored must
The test apparatus consists of three parts area. The displacement gauges may be also be surface-dressed to ensure that
(figure 1): dial gauges or have an analogue or digital rods, supposed to be fixed, are not set on
∑ A 2000-3000 kN hydraulic cylinder is ali- output, accurate to one-hundredth of a rock spalls, not intimately attached to the
gned with a stanchion of variable length millimetre. surrounding rock.
between two rigid plates. They must be One or two extensometers may be instal- As in the dilatometer test, the rigid plate
small (280mm diameter) in order to apply a led in small boreholes under the rigid bea- loading test proceeds by loading/unloa-
sufficiently high force to neutralise the ring plate to identify decompressed sur- ding cycles to increasing maxima, maintai-
most decompressed zone at the adit wall, face zones. ning load constant for short times in each

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 41

APPENDIX Caracterisation of rock masses useful for the design and the construction of underground structures

cycle to read the gauges and for a longer a semi-infinite, homogeneous, isotropic, The stress strain curves for successive
time at each cycle maximum to detect any elastic material with modulus of elasticity E cycles are plotted for each test (figure 2)
creep that might occur. The test may be and Poisson's ratio ν. The displacement of and so-called deformation global moduli
entirely automated, especially for creep the plate ∆d at each stress interval conside- are found, along with higher reversible
measurements. red is moduli. A global deformation modulus is
Test data is interpreted with the ∆d = π/4 x (1 - ν 2) x ∆σ x φ /E also determined, which corresponds to the
Boussinesq equation for a circular rigid If ν is not known, it is usually assigned a mean slope of the tangent to the cycle
plate of diameter f and a stress s applied to value of 0.25. curves.

Appendix 11 – Figure 2 – Curves from plate loading test and derived moduli

Appendix 11 – Figure 1 – Rigid plate loading test apparatus set-up

APPENDIX 12

THERMAL ROCK PARAMETERS

Rock Conductivity // Conductivity _

λ1 (W/m.°C) λ2 (W/m.°C)
Granites 2.7 – 3.5 2.7 – 3.5
Basalt 2.2 2.2
Appendix 12 – Table 1
Gneiss 3 – 4 2V6 - 3 Thermal conductivities of selected rocks (after
Mechler 1982, Handbook of Chemistry and Physics,
Crystalline schists 3.3 – 4.7 2.5 - 3 1987, Goy 1996)
Quartzites 5-5.6 5
Anhydrite 5.5 5.5 The lower the thermal conductivity value,
the more the rock acts as a heat insulant
Sandstone 2.7 2.7
Limestones 2.5-3.6 1.9-3.6
Dolomite 4.2 4.2
Clay 1.9 1.9
Rock salt 5.7 5.7

42 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 APPENDIX 13

BIENIAWSKI’S RMR CLASSIFICATION OF ROCK MASSES

Caracterisation of rock masses useful for the design and the construction of underground structures

Appendix 13 – Table 1 (cont’d) – Definition of RMR (after Bieniawski 1989)

APPENDIX
 APPENDIX

APPENDIX 13

BIENIAWSKI’S RMR CLASSIFICATION OF ROCK MASSES

Appendix 13 – Table 1 (cont’d) – Definition of RMR (after Bieniawski 1989)

Caracterisation of rock masses useful for the design and the construction of underground structures
 APPENDIX
Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX 14

BARTON’S Q INDEX

DESCRIPTION VALUE NOTES

ROCK QUALITY DESIGNATION RQD
Very poor 0-25 i) Where RQD is reported or measured
Poor 25-50 as < 10 (including 0) a nominal
Fair 50 - 75 value of 10 is used to evaluate Q
Good 75 - 90 ii) RQD intervals of 5, i.e. 100, 95,90 etc. are suffi-
ciently accurate
Excellent 90 -100
JOINT SET NUMBER Jn
Massive, no or few joints 0.5 – 1.0
One joint set 2
One joint set plus random 3
Two joint sets 4
Two joint sets plus random 6
Three joint sets 9 i) For intersections use (3.0 x Jn)
Three joint sets plus random 12
Four or more joint sets, random, heavily jointed,
“sugar cube”, etc 15 ii) For portals use (2.0 x Jn)
Crushed rock, earthlike 20
JOINT ROUGHNESS NUMBER Jr
a) Rock wall contact
b) Rock wall contact before 10 cm shear
Discontinuous joints 4
Rough and irregular, undulating 3
Smooth undulating 2
Slickensided undulating 1.5
Rough or irregular, planar 1.5
Smooth, planar 1.0
Slickensided, planar 0.5 i)Add 1.0 if the mean spacing of the relevant joint set
is greater than 3
c) No rock wall contact when sheared
Zones containing clay linerals thick enough to 1.0 ii) Jr = 0.5 can be used for planar, slickensided joints
prevent rock wall contact (nominal) having lineations,
Sandy, gravely or crushed zone thick enough o 1.0 Provided that the lineations are
prevent rock wall contact (nominal) oriented for minimum strength.
JOINT ALTERATION NUMBER Ja _r degrees (approx.)
a) Rock wall contact
Tightly healed, hard, non-softening, impermeable filling 0.75 i) Values of _r, the residual
angle friction
Unaltered joint walls, surface staining only 1.0 25 - 35 Are intended as an approximate
Slightly altered joint walls, non-softening mineral 2.0 25 - 30 Guide to the mineralogical properties
coatings, sandy particles, clay-free, disintegrated of the
rock, etc.
Silty, or sandy-clay coatings, small clay fraction 3.0 20 - 25 Alteration products, if present
(non-softening)
Softening or low-friction clay mineral coatings, i.e. 4.0 8 - 16
kaolinite, mica. Also chlorite, talc, gypsum and
graphite etc., and small quantities of swelling clays
(Discontinuous coatings 1 – 2 mm or less in thickness)

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 45

APPENDIX Caracterisation of rock masses useful for the design and the construction of underground structures
DESCRIPTION VALUE NOTES
JOINT ALTERATION NUMBER Ja _r degrees (approx.)
b) Rock wall contact before 10 cm shear
Sandy particles, clay free, disintegrating rock, etc. 4.0 25 -30
Stronly over-consolidated, non-softening clay minerals 6 16 - 24
fillings (continuous <5 mm thick)
Medium or low over-consolidation, softening clay 8 12 - 16
mineral fillings (continuous <5 mm thick)
Swelling clay fillings, i.e. montmorillonite, 8.0 –12.0 6 -12
(continuous <5 mm thick). Values of Ja depend on
percent of swelling clay-size particles, and access to water
c) No rock wallcontact when sheared
Zones or bands of disintegrated or crushed 6.0
Rock and clay 8.0
Zones or bands of silty-or sandy clay, small clay fraction, 5.0
non-softening
Thick continuous zones or bands of clay 10.0 – 13.0
JOINT WATER REDUCTION Jw Approx.water
pressure (kgf/cm2 )
Dry excavation or minir inflow i.e. < 5 l/m locally 1.0 < 1.0
Medium inflow or pressure, occasional outwash of joint fillings 0.66 1.0 – 2.5 i) Factors are crude estimates;
Large inflow or high pressure in competent rock with 0.5 2.5 – 10.0 increase Jw if drainage
installed
unfilled joints
Large inflow or high pressure 0.33 2.5 – 10.0
Exceptionally high inflow or pressure at blasting, 0.2 – 0.1 > 10 ii) Special problems caused
decaying with time
Exceptionally high inflow or pressure 0.1 – 0.05 > 10 By ice formation are not
considered

STRESS REDUCTION FACTOR SRF

a) Weakness zones intersecting excavation,
which may cause loosening of rock mass when
tunnel is excavated
Multiple occurrences of weakness zones containing 10.0 i)Reduce these values of SRF by 25-50% if the relevant shear
clay or chemically disintegrated rock, very loose zons only influence but do not intersect the excavation
surrounding rock and depth)
Single weakness zones containing clay, or chemically
disintegrated rock (excavation depth < 50 m) 5.0
Single weakness zones containing clay, or chemically
disintegrated rock (excavation depth > 50 m) 2.5
Multiple shear zones in competent rock (clay free),
loose surrounding rock (any depth) 7.5
Single shear zone in competent rock (clay free)
(depth of excavation <50 m) 5.0
Single shear zone in competent rock (clay free)
(depth of excavation >50 m) 2.5
Loose open joints, heavily jointed or “sugar cube”,
(any depth) 5
b) Competent rock, rock stress problems σc / σ1 σt / σ1 SFR
Low stress, near surface > 200 > 13 2,5 i) For strongly anisotropic virgin stress field
(if measured): when
Medium stress 200 - 10 13 –0.66 1.0 5<σ1 / σ3 <10 reduce σc to 0.8σc
High stress, very tight structure (usually favourable
to stability, may be unfavourable to wall stability) 10 - 5 0.66 – 0.33 0.5-2 and σt to 0,8σt.When σ1 / σ3 >10
Mild rockburst (massive rock) 5-2.5 0.33-0.16 5-10 Reduce σc and σt to 0.6σc and 0.6 σt
Heavy rockburst (massive rock) < 2.5 < 0.16 10-20
c) Squeezing rock plastic flow og incompetent
rock under influence of high rock pressure
Mild squeezing rock pressure 5 – 10 ii) Few case records
Heavy squeezing rock pressure 10 -20 avalaible where depth of crown below
d) Swelling rock, chemical swelling activity surface is less than span width. Suggest
depending on presence of water
Mild swelling rock pressure 5 – 10 SRF increase from
Heavy swelling rock pressure 10 -15 2.5 to 5 for such cases
 APPENDIX
Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX 15

EMPIRICAL FORMULAE FOR EVALUATING ROCK MASS DEFORMATION MODULI

Many empirical relationships have been ROCK MASS DEFORMATION MODULUS AUTHORS
proposed by various authors to estimate EM (MPa)
the rock mass deformation modulus from
the characteristics obtained from labora- 2*(RMR – 50) ou 1.7* (RQD –60) Cording et al (1971)
tory samples and other parameters. Table 25 Ln Q ou 10 Ln Q Fujita (1977)
1 lists the more important correlations with
0,7*(RMR/100)2 *Ei Barton (1980)
RMR, RQD and Q index used in rock mass
3 5
classification systems. RMR/10 + (RMR /10 ) Hoek & Brown (1982)
AFTES expresses no opinion on the rele- 10(RMR-10)/40 Serafim & Pereira (1983)
vance or validity of these correlations in 0,5*(RQD/100)2 *Ei Bieniawski (1989)
respect of the use to which they are put.
Users should refer to the writings of the 10*exp[(RMR-10)/40] Grimstad & Barton (1993)
relevant authors. 0,07*RQD+0.05 σc+55*Ei Hönisch (1993)
0,5 (GSI-10)/40
1000 *[σc/100] *10 (σc<100 MPa) Hoek & Brown (1997)
Appendix 15 – Table 1 – Empirical relationships proposed by various authors for assessing rock mass
deformation modulus

• Ei Modulus of elasticity of rock measured on laboratory • RMR Rock Mass Rating (Bieniawski 1989)
samples • RMR = 50 + 15 log10 Q (Barton 1995)
• σc Uniaxial compressive strength of rock measured on labora- • GSIGeological Strength Index
tory samples
• GSI = RMR89 – 5 (Hoek & Brown 1994, 1995); GSI = 9 Ln Q + 44
• RQD Rock Quality Designation (Deere 1967) (Bieniawski 1989)
• Q Quality factor (Barton 1980)

APPENDIX 16

REFERENCES

16.1 - BIBLIOGRAPHIE GENERALE HUDER J. & AMBERG G. 1970 - "Quellung in Mergel, Opalinuston
und Anhydrit", Schweizerische Bauzeitung n° 43, p. 975 – 980.
Comité Français de Mécanique des Roches (2000) - "Manuel de
Mécanique des Roches – Tome 1 – Fondement " – Coordonné par I.S.R.M. (1999) - "Suggested methods for laboratory testing of swel-
HOMAND F. & DUFFAUT P. – Presse de l’Ecole des Mines – Paris, 268 p. ling rocks " - International Journal of Rock Mechanics and Mining
Sciences, 36, p. 291 – 306.
BOUVARD-LECOANET A., COLOMBET G., ESTEULLE G. (1988) -
" Ouvrages souterrains-Conception-Réalisation-Entretien " - Presses I.S.R.M. (1985) - "Suggested methods for determining point load
de l’Ecole Nationale des Ponts et Chaussées, 262p. strength " - International Journal of Rock Mechanics and Mining
Sciences, 22 (2), 51-60.
HOEK E. & BROWN ET. (1980) - "Underground excavations in rock"
MOVINKEL T. & JOHANNESSEN O. (1986) - "Geological parameters
– London : Int. Min. Metall. 1980.
for hard rock tunnel boring." - Tunnels & Tunneling, april 86, pp. 45-48.
PANET M. & FOURMAINTRAUX D. (1976) – "La Mécanique des
roches appliquée aux ouvrages de génie civil ", Association Amicale 16.3 - CHARACTERISATION PARAMETERS OF
des Ingénieurs Anciens élèves de l’E.N.P.C., Paris, 236 p. DISCONTINUITIES
BARTON N. & CHOUBEY V. (1977) - "The shear strength of Rock
16.2 - ROCK MATRIX CHARACTERISATION Joints in Theory and Practice" - Rock Mech. Engng. Geol. 10,
PARAMETERS pp. 1-54.
BROCH et FRANKLIN, Int. J. of Rock Mech. And Min.Sc., 1972, DEERE D.U. (1963) - "Technical description of rock cores for enginee-
pp.669 – 697. ring purposes" - Rock Mech. Engng. Geol. 1, pp.16-22.

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 47

APPENDIX Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX 16

REFERENCES

LOUIS C. (1974) - "Reconnaissance des massifs rocheux par sondages Commission on Standardization of laboratory and field tests : Suggested
et classifications géotechniques des roches" - Sols et Fondations, methods for the description of discontinuities in rock masses" - Int. J.
108, N°319, pp. 97-122. Mech. Min. Sci. & Geomech. Abstr., Vol. 15, pp. 319-368.
SERRATRICE J.F. & DURVILLE J.L (1997) - "Description des roches et MILLER R.P. (1965) - "Engineering classification and index properties
des massifs rocheux : exploitation de deux bases de données" - for intact rock" - Ph.D. Thesis, University of Illinois.
Bulletin des Laboratoires des Ponts et Chaussées, 211, pp. 73-87.
16.6 - HYDRAULIC PARAMETERS
16.4 - SET PERSISTENCE OF DISCONTINUITIES (APPENDIX 10)
AND 3D GEOMETRICAL MODELLING OF BARTON N., BANDIS S. & BAKHTAR K.(1985) - "Strength, deforma-
NETWORKS OF DISCONTINUITIES tion and conductivity coupling of rock fractures" - Int. J. Mech. Min.
(APPENDIX 8) Sci. & Geomech. Abstr., Vol. 22, pp. 121-140.
BAECHER G.B., LANNEY N.A. & EINSTEIN H.H. (1977) - "Statistical BARTON.& DE QUADROS E.F., (1997) - "Joint aperture and rough-
distribution of rock properties and sampling" - Proc. 18th U.S. Symp. ness in the prediction of flow and groutability of rock masses" - Int. J.
on Rock Mech., Colorado, pp. 5c1.1-8. Mech. Min. Sci. & Geomech. Abstr., Vol. 34, No. 3-4, Paper
No.252.
BILLAUX D. (1990) - "Hydrogéologie des milieux fracturés : géomé-
trie, connectivité et comportement hydraulique" - Thèse de docto- GENTIER S. (1986) - "Morphologie et comportement hydroméca-
rat, CIG, Ecole des Mines de Paris. nique d'une fracture naturelle dans le granite sous contrainte normale;
étude expérimentale et théorique" - Documents du BRGM b34,
CACAS M.P. (1989) - "Développement d'un modèle tridimensionnel Orléans, 597 p.
stochastique discret pour la simulation de l'écoulement et des trans-
ferts de masse et de chaleur en milieu fracturé" - Thèse de doctorat, LOUIS C. (1969) - "Etudes des écoulements d'eau dans les roches fis-
surées et leur influence sur la stabilité des massifs rocheux" - BRGM,
CIG, Ecole des Mines de Paris.
Bulletin de la Direction des Etudes et Recherches. Série A.(3), pp.
HELIOT D. (1988) - "Conception et réalisation d’un outil intégré de 5-132.
modélisation des massifs rocheux fracturés en blocs" - Thèse de doc-
DE QUADROS E.F. (1982) - "Determinacão das caracteristicas+ de
torat INPL, Nancy. fluxo de agua em fracturas de rochas" - Dissert. de Maestrado.
HUDSON J.A. & PRIEST S.D. (1979) - "Discontinuities and rock mass Dept. of Civil Eng., Polytech. School, University of São Paulo.
geometry" - Int. J. Mech. Min. Sci. & Geomech. Abstr., Vol. 16, pp.
339-362. 16.7 - ROCK MASS CHARACTERISATION
LONG J.C.S, GILMOUR P. & WISERSPOON P.A. (1985) - "A PARAMETERS
model for steady fluid flow in random three dimensional networks 16.7.1 - Indirect measurements
of disk-shaped fractures" - Water resources research, 21, N°8,
BOUVARD A., HUGONIN J. & SCHNEIDER, B. (1994) - "SCARA-
pp. 1105-1115. BEE, - Méthode de reconnaissance des massifs rocheux – Application
PAHL P.J. (1981) - "Estimating the mean length of discontinuity traces" aux ouvrages souterrains" – Tunnels et ouvrages souterrains n°123
- Int. J. Mech. Min. Sci. & Geomech. Abstr., Vol. 18, pp. 221-228. Mai/Juin 1994.
WARBURTON P.M. (1980) - "Stereological interpretation of joint trace CARRERE A., RIVET, J. & SCHNEIDER B. - "La petite sismique" –
data, influence of joint shape and implication for geological surveys" - Géologues n° 92.
Int. J. Mech. Min. Sci. & Geomech. Abstr., Vol. 17, pp. 305-316. 16.7.2 - Rigid plate loading test
XU J. (1991) - "Simulation statistique de discontinuités et évaluation de MAZENOT P. - "L’essai à la plaque – Congrès des Grands Barrages –
la blocométrie de massif rocheux, application à l’analyse de l’écoulement Edimbourg – 1964 – Q.28 – Rapport 15 – Groupe de travail". –
et de la stabilité" - Thèse de doctorat, CGI, Ecole des Mines de Paris. Ann.ITBTP, 1965.
16.7.3 - Stresses
16.5 - SCHMIDT HAMMER TEST (APPENDIX 9)
AMADEI B. & STEPHANSON O. (1997) – “Rock Stress and its
BARTON N. & CHOUBEY V. (1977) - "The shear strength of Rock Measurement” – Chapman and Hall, London, 490 p.
Joints in Theory and Practice" - Rock Mech. Engng. Geol. 10,
BRIGLIA P., BURLET D. & PIRAUD J. (1994) – “La mesure des
pp. 1-54.
contraintes naturelles appliquée au Génie civil ” – Tunnel et
HOEK E. & BRAY J.W. (1974) - "Rock slope Engineering" - The Ouvrages Souterrains, n°123, mai-juin 1994, pp. 149-152.
Institution of Mining and Metallurgy, London, revised third edition BURLET D. (1991) – “ Détermination du champ des contraintes régio-
1981. nal à partir de tests hydrauliques en forage ”. Thèse de doctorat,
I.S.R.M. (1978) - "International Society for Rock Mechanics Univ. Paris VII.

48 TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003

 APPENDIX
Caracterisation of rock masses useful for the design and the construction of underground structures

APPENDIX 16

REFERENCES

Comité Français de Mécanique des Roches de l'exhaure en mines et carrières" Mines et NF P 94-410-2 (Mai 2001) – "Essais pour
(2000) - "Manuel de Mécanique des Roches Carrières - Industrie Minérale. Février 1992. déterminer les propriétés physiques des
– Tome 1 – Fondement- Chap. 7 – Les roches – Partie 2 : Détermination de la masse
LOMBARDI G. "Hydrogeologische Aspekte
contraintes dans les massifs rocheux et leur volumique – Méthode géométrique et par
von Tunnelprojekten" Felsbau. Vol. 12, n° 6,
détermination " –Coordonné par immersion dans l’eau" – AFNOR 2001
1994.
HOMAND, F. et DUFFAUT, P.– Presse de
l’Ecole des Mines – Paris, 268 p. GAUDIN B. "Maîtrise de l'eau dans les tun- NF P 94-410-3 (Mai 2001) – "Essais pour
nels en construction et en service". Conf. à déterminer les propriétés physiques des
FABRE D., MAYEUR B. & SIRIEYS,P. (2002) -
l'Ecole des Ponts et Chaussées, 27 mars roches – Partie 3 : Détermination de la poro-
"Paramètres caractérisant l’état de
1997. sité" – AFNOR 2001
contraintes naturel dans les massifs
rocheux". - Symposium "Paramètres de cal- NF P 94-411 (Avril 2002) – "Détermination
cul géotechnique" – Magnan (ed), Presses 16.8 - USE OF ROCK MASS de la vitesse de propagation des ondes ultra-
de l’E.N.P.C., Paris p. 359-368. CHARACTERISATION FOR sonores en laboratoire – Méthode par trans-
DESIGN AND CONSTRUC- parence" – AFNOR 2002
16.7.4 - Thermal rock parameters TION OF UNDERGROUND NF P 94-420 (Décembre 2000) –
(appendix 12) STRUCTURES
FABRE D. (2001) - "Prediction of tem- "Détermination de la résistance à la com-
perature for deep tunnel projects". Pro. Fascicule Génie civil Dossier pilote des pression uniaxiale" – AFNOR 2000
Int. Symp.Geonics – Temperature and Tunnels – Génie Civil (Juillet1998)
its influence on Geomaterials – Ostrava, NF P 94-422 (Janvier 2001) – "Détermination
pp. 67 –75. Section 1 : Introduction de la résistance à la traction – Méthode indi-
Section 2 : Géologie – Hydrogéologie – recte – Essai brésilien" – AFNOR 2001
GABLE R. (1986) - "Température, gradient et
Géotechnique NF P 94-423 (Mars 2002) – " Détermination
flux de chaleur terrestre. Mesures, interpréta-
Section 3 : Conception et dimensionnement de la résistance à la compression triaxiale" –
tion" – Document du BRGM, n° 104, 187 p.
Centre d’Etudes des Tunnels. AFNOR 2002
GOY L. (1996) - "Mesure et modélisation
des températures dans les massifs rocheux. HOEK E. & BROWN ET. – "Practical esti- NF P 94-425 (Avril 2002) – " Détermination
Application au tunnel profond Maurienne- mates of rock mass strength" – Int.J.Rock du module de Young et du coefficient de
Ambin." – Thèse de doctorat Mech. Mn-Sci, Vol 34, pp.1165-1186, 1997 Poisson" – AFNOR 2002
Géomécanique, Univ. Grenoble, 202 p. HOEK E., KAISER P.K. & BAWDEN W.F. - NF P 94-430-1 (Octobre 2000) – "Détermination
MECHLER P. (1982) - "Les méthodes de la "Support of underground excavations in du pouvoir abrasif d’une roche – Partie 1 :
géophysique" – Dunod éd. Paris 2000 hard rock" – Rotterdam, Balkema. 1995 Essai de rayure avec une pointe" – AFNOR
2000
HODGMAM Ch.D. (1987) - "Handbook of
Chemistry and Physics" – 30ème édition - NF P 94-430-2 (Octobre 2000) – "Détermination
Evaluation de la résistance limite des mas-
Chemical Rubber Publishing Company, du pouvoir abrasif d’une roche – Partie 1 :
sifs rocheux
Cleveland , Ohio, USA, 2651 p. Essai avec outil en rotation" – AFNOR 2000
BALMER G. "A general analytical solution
FABRE D., GOY L., MENARD G. & BURLET NF P 94-429 (en cours d’établissement) –
for Mohr's enveloppe" - Am. Soc. Test. Mat.
D. (1996) - "Température et contraintes dans "Détermination de la résistance à la com-
52, 1260-1271
les massifs rocheux : cas du projet de tunnel pression d’une roche entre pointes – Essai
Maurienne-Ambin" – Tunnels et ouvrages Franklin" – AFNOR 2000
16.9 - AFNOR STANDARDS
souterrains, n° 134, p.85 – 92. P 18-572 (Décembre 1990) – " Essai d’usure
NF P 94-066 (Décembre 1992) – micro-Deval" – AFNOR 1990
16.7.5 - Hydrogeological conditions "Coefficient de fragmentabilité des maté-
P 18-573 (Décembre 1990) – " Essai Los
BORDET Cl. "L'eau dans les massifs rocheux riaux rocheux" – AFNOR 1992
Angeles" – AFNOR 1990
fissurés ; observations dans les travaux sou-
terrains". Conf. au CERES, Liège, 14 NF P 94-067 (Décembre 1992) – "Coefficient XP P 94-443-1 (Février 2002) –
décembre 1970. de dégradabilité des matériaux rocheux" – "Déformabilité – Essai dilatométrique en
DUFFAUT P. "Les tunnels en terrain aqui- AFNOR 1992 forage – Partie 1 : Essai avec cycles" –
fère" In : La pratique des sols et fondations, NF P 94-410-1 (Mai 2001) – "Essais pour AFNOR 2002
coord. G. FILLIAT, Ed. du Moniteur, 1981, déterminer les propriétés physiques des XP P 94-443-2 (Février 2002) –
pp. 808-810. roches – Partie 1 : Détermination de la teneur "Déformabilité – Essai dilatométrique en
VANDENBEUSCH M. & WOJTKOWIAK F. en eau pondérale – Méthode par étuvage" – forage – Partie 1 : Essai de fluage après le
"La prise en compte de l'eau et la prévision AFNOR 2001 premier cycle" – AFNOR 2002

TUNNELS ET OUVRAGES SOUTERRAINS - N° 177 - MAI/JUIN 2003 49