C.

Derek Martin
University of Alberta, Edmonton, Canada

Why measure in-situ stress?

ƒ Engineering analyses require boundary

conditions
ƒ One of the most important boundary
conditions for the analysis of underground
excavations is in-
in-situ stress

σ1 Failure
Strength
FOS= Stress

In Situ Stress State

σ3

1
What is in-situ stress?
F3 Boundary
F2
conditions
F1
Fn
Intact rock Excavation

Water flow

Discontinuities

Total Stresses = In situ stress + Excavation-Induced stress

Definitions

Stress at a point
σzz
Normal Shear σyy
Stress (σ) Stress (τ) τzx
τzy
τyz τxy
σxx σxx
τxz
τyx
σyy

σzz

The normal and shear forces acting at a point in the

rock mass are represented by three normal and six
shear stresses.

2
Definitions (cont’d)

Normal & Shear Stresses Principal

3-dimensions Stresses
σ xx τ xy τ xz  σ xx τ xy τ xz  σ 1 0 0 
  
 τ yx σ yy τ yz   σ yy τ yz  
 σ 2 0 
 τ zx τ zy σ zz  Symmetrical σ zz   σ 3 
 
The components in a row are the
Planes of Principal
components acting on a plane;
stress have no
for row 1, the plane on which σxx acts.
shear stresses.
Example:
The components in a column are the
Excavation surface
components acting in one direction;
for column 1, the x direction.

Co-ordinate systems

Geotechnical Engineering
-Right-hand systems
-Compression positive
-Tension negative

3
Data presentation

Measurements

σ¾11/T r/Pl σ¾22/T r/Pl σ¾33/T r/Pl

Borehole
Depth (m)
E
(MP a)
ν
º
MPa/ °±/ ±° MPa/ ±°/ ±° MPa/ °±/ °±
12.87 49.1 0.12 33.1/237/25 18.9/339/23 16.2/106/55
13.43 47.5 0.14 26.3/238/17 14.5/136/36 7.9/349/ 49
13.94 50.3 0.15 33.1/233/29 17.0/142/02 14.7/049/61
14.50 51.2 0.14 34.1/244/09 18.8/145/45 13.0/324/44

Magnitudes and Orientation

Orientation Data
Magnitude Data
Stress (MPa) Trend Plunge
(º) (º)
Sigma 1 10 210 70
Sigma 2 8 320 10
σ 1 0 0  Sigma 3 5 50 15
 σ 2 0 

 σ 3 

Need to now the in-situ stress

in the plane of a tunnel for
plane strain analysis

4
Useful stress equations

Kirsch equations

Stress ratio
Max. Tangential Stress
σθmax=3σ1-σ3 = 3(k-1)σ3 k = σ1/σ3

Min. Tangential Stress

σθmin=3σ3-σ1= 3(1-k)σ3

Stress path

Stress path

5
The World Stress Map Project

Web Site: http://www-

http://www-wsm.
wsm.physik.
physik.uni-
uni-karlsruhe.de/
karlsruhe.de/

Types of Data:
ƒ earthquake focal mechanisms (63%)
ƒ well bore breakouts and drilling induced
fractures (23%)
ƒ in-
in-situ stress measurements (overcoring
(overcoring,,
hydraulic fracturing, borehole slotter (9%)
ƒ young geologic data (from fault slip
analysis and volcanic vent alignments (5%)

Europe

6
The Alps

Role of stress data bases for new projects

Can we rely on these stress data bases – Canadian Example

Compiled by CANMET:Measurements to 2500 m

7
North American Stresses

Site C

URL
Sudbury

CANMET Stress database

Sudbury Basin Measurements

Onaping Depth
2500 m ?

8
CANMET Database: Magnitudes

Canadian Shield

Sigma 1 Sigma 2

Possible stress ratios

Linear Fit Nonlinear Fit

σv σv
σH σH

Normal Faults Thrust Faults

σv > σH σH > σv

9
Sweden

Aspö HRL

Sweden’s stress database

Sigma 1 (MPa) Sigma 2 (MPa) 0

Sigma
20 40
3 (MPa)
60 80
0 20 40 0 10 20 30 40 50
0
0 0

100
100 100

200 200 200

300 300 300

400 400 400

500 500 500

600 600 600

700 700 700

800 800 800

10
SKB’s Äspö Hard Rock Laboratory

Stress & water-bearing Fractures

σ1

11
Stress & Subhorizontal fractures
Aspo HRL Forsmark URL (Canada)

Stress magnitudes versus depth

Summary from detailed measurements

12
AECL’s URL Canada

Stress orientations & Geological structure

13
What about σ3

Predicting Stress

Sigma 3
Horizontal Stress

Displacement
Boundary
condition

14
Weak rocks – Western Canada

Cretaceous Clay Shales – Marine origin

Thrust Fault

Site C Stresses (Canada)

Plan Plan

Section

15
Nagra’s Mont Terri Lab

Mont Terri – Opalinus Clay Shale

16
In-Situ Stress Tensor Measurements

Rosas1
Stress (MPa) Trend Plunge
(º) (º)
Sigma 1 6-7 210 70
Sigma 2 4-5 320 10
Sigma 3 0.6-2 50 15

Gives σn=4.6 MPa acting

perpendicular to the bedding

Slotter
Stress (MPa) Trend Plunge
(º) (º)
Sigma 1 3.1 181 45
Sigma 2 1.6 074 16
Sigma 3 0.1 330 0.17

Gives σn=0.4 MPa acting

perpendicular to the bedding

Topography and σ3

Valley above the Rock Lab cuts down to the

Southwest to elevations 100m below the
Rock Lab at St Ursanne.
σ3 Distance along profile
Negative values to South of Rock Lab
1000
Elevation (masl)

750

500

250
-2000 -1500 -1000 -500 0 500 1000 1500 2000
Distance Along Profile (m)
Elevation

200

36° 0

-1000 0

Easting
Elevation

200

-1000 0 1000

Northing

17
Opalinus Clay (Geological Structures)

In-situ Stress
at Mont Terri

Factors affecting in situ stress

1. Does geological structures affect in situ stresses?

YES
2. Do different measurement methods used yield
comparable in situ stress fields? Sometimes
3. What is the importance of residual stresses, i.e.
stresses locked into the rock mass?
For engineering not important

A C
B

Fracture

Rock mass

18
Stress modelling

Stress modelling at Aspo HRL

a) b)

EW-1b
c) d)

NE2
I EW3
EW-1a
EW3 NE2 L

NE1
EW1a
NE1
G H
EW1b N

Measurement methods

ƒ Overcoring
• Triaxial Strain Cells
• Borehole Deformation gauge
• Slotter
ƒ Fracturing
• Hydraulic Fracturing
• Sleeve Fracturing
ƒ Undercoring
ƒ Convergence

19
Undercoring

ƒ Developed at AECL’s URL when traditiional methods

failed
• Originally used 8 CSIRO cells

ƒ Involves a large rock-

rock-mass volume
ƒ Used when traditional overcoring is affected by:
• Micro-cracking and nonlinear strains
• Core disking

ƒ Expensive method requiring sufficient redundancy to

constrain the solution
• Measure strain, such as CSIRO cells and
• Determine the stress tensor that produces the strains
using a 3D Numerical Program, e.g., Flac3D
• Depends on the small strain region to determine the tensor
hence redundancy is quite important.

Undercore method at Mont Terri

20
Effect of Scale

Residual strain
AECL Residual Strain Experiments

The residual strain measured was <5% of the stress magnitudes.

Not of engineering significance at the URL

21
Tunnel convergence measurements

Convergence and Stress

Assumptions:
-Elastic response
-Estimate of response ahead of tunnel

22
Observations

Evidence for stress orientations

ƒ Borehole breakouts
ƒ Core disking
ƒ Stress fracturing around tunnels

Magnitudes can be constrained but no absolute

values can be given at this time!

Borehole Breakouts

300-mm-diameter borehole

23
Breakouts: BDT4
Rosas1 Stress Tensor

Note: Borehole influenced by tunnel

Borehole Breakouts: 75-

75-mm diameter

24
Breakout examples

No Breakout Irregular Classic

Breakout Breakout

Core Disking

Region of
Stress σmin
concentrations

σmax

Core
disking

25
Core disking & Scale

1.2-m-diameter

300-mm-diameter

Same Stress state

Conclusion:
Core disking
Independent of scale
76-mm-diameter

Lessons from stress measurement programs

ƒ Stress databases only provide trends

ƒ Stress must be measured if stress-related
consequences important.
ƒ Requires specialists who are knowledgable with
various techniques.
ƒ Large-scale measurements (Undercore) preferred
over small-scale measurements such as overcoring
ƒ When using Overcoring
• Glued triaxial strain cells for hard rocks
• Borehole deformation gauge for soft rocks
• When σ1>0.2σc limit for elastic behaviour is close for
crystalline rocks
ƒ Hydraulic fracturing for deep boreholes (500 m
overcore tests now possible)

Do not discard results (without justification) – they might

be valid

26