Optimising Geotechnical Logging to

Accurately Represent the Geotechnical

Environment
G D Dempers1, C R W Seymour2 and M B Harris3

ABSTRACT
Typically, large volumes of data are collected during geotechnical investigations for mining projects
but rarely is this done in a comprehensive manner that enables all aspects of the geotechnical
environment to be evaluated.
A geotechnical core logging process has been developed to record mechanical and structural
properties of the rock mass. The method enables data for a wide range of rock properties and
geotechnically significant major structures to be collected including rock strength, joint surface
condition, fracture frequency and fracture orientation. The logging method is unique in that
sufficient data is collected to enable the independent determination of all the major rock mass
classification systems including rock mass rating (RMR), (Bieniawski 1976, 1989; Laubscher,
1990) Norwegian Geotechnical Institute (NGI) tunnelling quality index (Q) (Barton, Lien and
Lunde, 1974) and geological strength index (GSI) (Hoek, Kaiser and Bawden, 1995).
The logging system has been specifically developed to allow better and more precise appreciation
of rock mass and structural conditions across the project area thereby optimising the use and
application of the available geotechnical data and improving confidence in the outcomes of
geotechnical investigations.
This paper describes the core logging process with case examples showing how the logged data
can be used for underground and open pit mine design.

INTRODUCTION
Large volumes of geotechnical data are often collected during through creating mining rock mass models for a wide variety
the life of a mining project but this data is rarely ever effectively of different projects that make this a rigorous, robust and
and comprehensively collected to represent the rock mass unique method. It has been used with equal success in many
conditions across the project. The purpose of geotechnical different environments and for both open pit and underground
core logging is to get an appreciation of the rock conditions projects. The geotechnical domains are determined by
and apply these understandings to mine design. grouping together rock which displays similar geotechnical
There are many different forms of geotechnical core logging characteristics and which will behave uniformly in an
data collection that range from established methods through to excavation. This domain logging allows the variability of rock
project/outcome specific templates developed ‘in-house’. The mass conditions within and across individual lithological/
format and legend for logging that is presented in this paper geological/structural units to be identified more readily than
has evolved with development of the mining rock mass model fixed interval methods with logging per metre or per drill run.
(Seymour, Dempers and Jenkins, 2007; Jenkins, Dempers As such, a domain can be many metres in length or less
and Seymour, 2009). It differs from other logging schemes in than one metre and is determined from significant lithological
that it is designed to help highlight rock mass variability and boundaries which are then further subdivided according to
in particular those regions or conditions in the rock mass that geological structure, weathering, hydrogeology, veining and
are likely to be problematic for design purposes. alteration within those major lithological boundaries.
The core logging method developed requires the core to be After the rock has been grouped into geotechnical domains,
grouped into logging intervals that are unique geotechnical each relevant parameter required for geotechnical evaluation
domains or designs regions within a particular rock type. is then logged within a particular geotechnical domain.
Most of the logged parameters and definitions have been Selected parameters include rock strength, discontinuity
used in other logging and rating schemes. However, it is the condition, rock quality designation (RQD), discontinuity
dedicated use of the logging technique and it’s refinement count per fracture angle and discontinuity orientation. The

1. MAusIMM, Principal, Dempers and Seymour Pty Ltd, Geotechnical and Mining Consultants, PO Box 2323, Warwick WA 6024. Email: [email protected]
2. MAusIMM, Principal, Dempers and Seymour Pty Ltd, Geotechnical and Mining Consultants, PO Box 2323, Warwick WA 6024. Email: [email protected]
3. Geotechnical Engineer, Dempers and Seymour Pty Ltd, Geotechnical and Mining Consultants, PO Box 2323, Warwick WA 6024. Email: [email protected]

SECOND AUSTRALASIAN GROUND CONTROL IN MINING CONFERENCE / SYDNEY, NSW, 23 - 24 NOVEMBER 2010 85
SECOND AUSTRALASIAN GROUND CONTROL IN MINING CONFERENCE / SYDNEY, NSW, 23 - 24 NOVEMBER 2010

Matrix Micro Macro Infill Micro Macro Infill

Weath QSI Fract Infill Infill JWA Fract Infill Infill JWA
Hole-id From To Rock Type RQD (m) Struct No.Sets (1-9) (1-5) (1-9) (1-9) (1-5) (1-9) Comments
(1-5) (1-5) 0-30 type Thick (1-3) 0-90 type Thick (1-3)
Type 0-30 0-30 0-30 0-90 0-90 0-90

MWRD142 50.0 65.0 tuff 3.5 2.5 2.10 3.5 250 3 2 5 iox/cly 1 NQ core ~3.0m loss mech broken/crushed i/p fol/lay i/p
MWRD142 65.0 65.7 tuff 3 3 0.50 2.5 7 7 2 5 iox/cly 1 fol/lay
MWRD142 65.7 66.5 tuff 2.5 3 0.60 2.5 6 5 2 5 iox/cly 1 fol/lay
MWRD142 66.5 69.6 tuff 2.5 2.5 1.30 3 45 3 2 5 iox/cly 1
MWRD142 69.6 71.8 tuff 3.5 1.5 0.01 m3 4.5 100 5 2 5 iox/cly 1 fol/lay ~ 0.4m loss mech broken/crushed i/p
MWRD142 71.8 73.2 tuff 2.5 2.5 1.00 2.5 18 5 2 5 iox/cly 1 fol/lay

OPTIMISING GEOTECHNICAL LOGGING TO ACCURATELY REPRESENT THE GEOTECHNICAL ENVIRONMENT

MWRD142 73.2 81.5 tuff 2 3 4.90 2.5 80 2 2 4 iox/cly 1 fol/lay
MWRD142 81.5 83.2 tuff 2.5 2 0.40 3.5 50 4 2 5 iox/cly 1 fol/lay mech broken i/p
MWRD142 83.2 86.2 tuff 2.5 2.5 1.60 2.5 28 2 2 6 1 fol/lay
MWRD142 86.2 86.5 tuff 3 1 0.01 m2 4.5 250 4 2 5 iox/cly 1 sheared
MWRD142 86.5 88.0 tuff 2.5 2.5 0.90 2.5 14 7 2 6 1 fol/lay
MWRD142 88.0 88.1 tuff 3.5 1 0.01 m3 4.5 100 5 2 5 iox/cly 1
MWRD142 88.1 90.5 tuff 2.5 2.5 1.00 2.5 38 4 2 5 iox/cly 1 fol/lay i/p
MWRD142 90.5 91.2 tuff 2.5 1.5 0.01 m3 4.5 100 5 2 5 iox/cly 1 1 ~0.1m loss
MWRD142 91.2 92.7 tuff 2 3 1.30 2.5 13 5 2 5 iox 1 fol/lay
MWRD142 92.7 93.5 tuff 2 2.5 0.20 2.5 13 8 2 5 iox 1 highly fol/lay
MWRD142 93.5 95.5 tuff 2 3 2.00 2.5 6 5 2 5 iox 1 highly fol/lay
MWRD142 95.5 95.6 tuff 2 1 0.01 m2 4.5 50 4 2 4 cly 2 1 sheared
MWRD142 95.6 97.3 tuff 2 3 1.40 2.5 11 4 2 5 iox 1 fol/lay
MWRD142 97.3 98.4 tuff 2 0.5 0.01 m2 5 250 1 2 3 cly 5 1 fol/lay/sheared ~0 2m loss
MWRD142 98.4 98.8 tuff 2 1.5 0.01 2.5 12 1 2 4 iox/cly 1 fol/lay
MWRD142 98.8 99.7 tuff 2 3 0.80 2.5 8 4 2 6 1 fol/lay
MWRD142 99.7 101.2 ?mafic tuff 2 2 0.30 m2 3.5 75 4 2 5 iox/cly 1 fol/lay/sheared core split from 101 0 m
MWRD142 101.2 101.8 ?mafic tuff/?dol 2 1 0.01 m2 4.5 100 1 2 5 iox/cly 1 sheared ~0.3m loss ? Dol contact zone Jw zone split
MWRD142 101.8 102.6 ?dol 1.5 2 0.30 2.5 10 6 2 7 iox 1 split
MWRD142 102.6 104.8 chlor hem 1 3 2.10 2.5 5 8 2 5 iox/cly 1 split
MWRD142 104.8 106.6 chlor tuff 1 1 0.10 m2 4.5 100 1 2 5 chlor 1 split micro-fractured sheared ~0.4m loss
MWRD142 106.6 112.6 hem/chlor hem 1 3 5.70 2.5 18 8 2 5 cly 1 split
MWRD142 112.6 113.7 chlor/hem tuff 1 1 0.01 m2 4.5 100 7 2 5 cly/chlor 1 split micro-fractured ~0.2m loss
MWRD142 113.7 115.6 hem/chlor hem 1 3 1.80 2.5 5 8 2 8 1 split micro-fractured i/p
MWRD142 115.6 117.8 chlor tuff 1 1 0.01 m2 4.5 100 1 2 6 chlor 1 split micro-fractured ~1.0m loss
MWRD142 117.8 121.4 tuff 1 3 3.50 2.5 8 2 2 5 iox 1 split to 119 0m ~massive Jw zone
MWRD142 121.4 121.7 tuff 2 2.5 0.10 2.5 6 5 2 5 iox 1 1 ~massive Jw zone
MWRD142 121.7 122.0 tuff 2.5 1.5 0.01 m3 4.5 50 2 2 5 iox 2 1 Jw zone
MWRD142 122.0 123.1 tuff 2 2.5 0.20 2.5 12 5 2 5 iox 1 ~massive Jw zone
MWRD142 123.1 129.5 tuff 1 3 6.10 2.5 26 2 2 5 iox 1 1 127.1m: 1X 10mm cly/fault gouge infilled

FIG 3 - Typical rock mass log (note 30° - 60° and 60° - 90° log not shown).
87
G D DEMPERS, C R W SEYMOUR AND M B HARRIS

1. unweathered, such, they are not recorded for every logged interval but are
2. slightly weathered, used sparingly for exceptional circumstances. The matrix
3. moderately weathered, codes most commonly used are:
4. highly weathered, and  M1 – fault (discrete),
5. completely weathered.  M2 – shear zone,
Quality strength index  M3 – intense fracturing,
The quality strength index (QSI) reflects the average?  M4 – intense mineralisation (usually ore),
estimated rock strength within a geotechnical domain. The  M5 – deformable material,
logged QSI range is from extremely weak (0.5) to extremely  M6 – discing (record metres in the comments column),
strong (five). RMR ratings and equivalent uniaxial compressive and
strengths (UCS) for possible logged values are shown in
 M7 – vuggy.
Table 1. Various published field estimates for UCS can be used
Additional project specific matrix codes may be employed
to help determine these ratings, for example Hoek, Kaiser and
where appropriate to indicate specific types of pervasive
Bawden (1995).
alteration, veining or other unique geotechnical features
Rock quality designation within the rock mass.
Rock quality designation (RQD) is the percentage of the
drilled length of a geotechnical domain which has recovered Joint sets
core lengths of 10 cm or greater. ‘Core 10’ is the total length This is the number of joint sets present within a geotechnical
of core which is greater than 10 cm within a geotechnical domain. The Q classification rating number associated with
domain which is measured and recorded in the log. RQD is the logged number of joint sets is given in Table 2. Logged
later calculated according to the following: values are generally 2.5 or greater if the logged intervals reflect
geotechnical domains at the engineering scale. That is, at the
scale of a tunnel wall or batter slope, rather than of intact rock
Total length of core > 10 cm
RQD % # 100 blocks between joints.
Length of geotechnical domain

Matrix and structure codes Fractures per interval

Matrix and structure codes are additional descriptors used This is the counted number of fractures per fracture angle
to help highlight conditions of geotechnical significance. As grouping, or ‘bin’ within the logged interval. Fracture angle bins

TABLE 1
Quality strength index.
Description Logged value RMR† rating Equivalent UCS (MPa) UCS range (MPa)
Extremely weak 0.5 1 1 <1
Very weak 1.0 1 4 1-5
Weak 2.0 3 25 5 - 25
Moderately strong 2.5 6 64 25 - 65
Strong 3.0 10 100 66 - 105
Strong to very strong 3.5 13 134 106 - 140
Very strong 4.0 15 154 141 - 160
Very hard to extremely strong 4.5 17 174 161 - 185
Extremely strong 5.0 18 185 >185
†
IRS strength rating for RMR after Laubscher (1990).

TABLE 2
Logged number of joint sets and Jn rating.
Description Logged values Q joint set number rating (Jn)
Massive or few joints 0.5 1
One joint set 1.0 2
One joint set plus random joints(s) 1.5 3
Two joint sets 2.0 4
Two joint sets plus random joints(s) 2.5 6
Three joint sets 3.0 9
Three joint sets plus random joints(s) 3.5 12
Four or more joint sets, random, heavily jointed 4.0 15
Crushed rock, earth-like 5.0 20

88 SECOND AUSTRALASIAN GROUND CONTROL IN MINING CONFERENCE / SYDNEY, NSW, 23 - 24 NOVEMBER 2010
G D DEMPERS, C R W SEYMOUR AND M B HARRIS

the large-scale joint surface characteristics at the scale of Joint condition (Jc): 0 - 40 (40 × micro × macro × infill
exposure (several metres). For logging purposes and unless × JWA from Tables 4,5, 6 and 8)
the specific joint condition is known, a macro-roughness Q is calculated as follows:
default descriptor of ‘undulating’ (logged value of two) is
applied.
Q = RQD%/Jn × Jr/Ja × Jw/SRF
When recording joint infill characteristics, slight traces of
infill material (not continuous over the fracture surface) that
where:
do not influence the shear strength/cohesion of the structure,
are for geotechnical purposes not considered as infill and the RQD = rock quality designation
structure is recorded as ‘clean’. Fracture infill and alteration Jn = joint set number (Table 2)
codes are based on the Q Index and RMR classification systems.
Joint wall alteration is logged according to the effect of Jr = joint roughness number (Table 4)
alteration on the wall rock of the joint as per Table 8. Ja = joint alteration number (Table 6)
Relevant logged values for each joint characteristic are Jw = joint water reduction factor, assumed to be 1 (0.1 - 1)
shown in Table 4 to Table 8 and where appropriate, equivalent
classification ratings are also provided. SRF = stress reduction factor (0.5 - 20)
Based on the quantitative approach (Cai et al 2004), GSI can
INTERPRETATION OF GEOTECHNICAL be determined from block size (joint spacing) and the joint
LOGGING condition factor defined as follows:
The raw logging data can be used to calculate various
geotechnical parameters and rock mass rating values, for Jc = Jw × Js/Ja
example RMR (Laubscher, 1990), rock tunnel quality index,
Q (Barton, Lien and Lunde, 1974) and GSI (Hoek, Kaiser and where:
Bawden, 1995). RMR is calculated as follows:
Jw = large-scale waviness determined from logged macro
RMR = FF + IRS + Jc joint roughness (Table 5)
Js = small scale smoothness determined from logged
where: micro joint roughness (Table 4)
Input parameter Rating range Ja = joint alteration determined from logged infill
Intact rock strength (IRS): 0 - 20 (Table 1) condition (Table 6)
Fracture frequency per 0 - 40 (Table 3) The rock mass can be classified according to GSI as shown
joint set (FF): in Figure 5.

TABLE 4
Micro joint roughness.

Description Logged value RMR rating RMR rating Q joint roughness GSI rating JRC
(moderate water) (dry) rating (Jr) (Js)
Polished or slickensided 1 0.45 0.55 0.5 0.5 0.5
and planar
Smooth and planar 2 0.50 0.60 1.0 1.0 1.0
Rough and planar 3 0.55 0.65 1.5 1.0 2.0
Slickensided undulating 4 0.60 0.70 1.5 1.5 3.0
Smooth undulating 5 0.65 0.75 2.0 1.5 4.0
Rough undulating 6 0.70 0.80 3.0 2.0 5.0
Slickensided stepped 7 0.75 0.85 3.0 2.0 6.0
Smooth stepped 8 0.80 0.90 3.0 3.0 7.0
Rough stepped 9 0.85 0.95 3.0 3.0 8.0

TABLE 5
Macro joint roughness.
Description Logged value RMR rating (moderate water) RMR rating (dry) GSI rating (Jw)
Planar 1 0.65 0.75 1.0
Undulating 2 0.75 0.80 1.5
Curved 3 0.75 0.85 2.0
Irregular, unidirectional 4 0.85 0.95 2.5
Irregular multi-directional 5 0.95 1.00 3.0

90 SECOND AUSTRALASIAN GROUND CONTROL IN MINING CONFERENCE / SYDNEY, NSW, 23 - 24 NOVEMBER 2010
 OPTIMISING GEOTECHNICAL LOGGING TO ACCURATELY REPRESENT THE GEOTECHNICAL ENVIRONMENT

TABLE 6
Joint infill condition.
Description Logged value RMR rating RMR rating Q joint alteration GSI rating
(moderate water) (dry) rating (Ja) (Ja)
Gouge >amplitude† 1 0.15 0.30 12 12
†
Gouge <amplitude 2 0.35 0.45 8 6
Soft sheared – fine 3 0.40 0.50 4 4
Soft sheared – medium 4 0.50 0.60 4 4
Soft sheared – coarse 5 0.60 0.70 4 4
Non-softening – fine 6 0.70 0.80 3 3
Non-softening – medium 7 0.75 0.85 3 2
Non-softening – coarse 8 0.80 0.90 2 1
Clean/surface staining 9 0.90 1.00 1 1
†
Gouge thickness greater or less than the amplitude of joint surface irregularities.
The rock strength is classified as very good (UCS >160 MPa)
TABLE 7 in Figure 4 but is poor to fair (Jr/Ja <2) in terms of joint shear
Infill thickness. strength (Figure 8).

Description Logged value CONCLUSIONS

Clean, no infill or insignificant 0 The development of the domaining logging system has
Thickness of infill <1 mm 1 enabled the transfer of an accurate representation of the rock
mass and structure to be applied to block, numerical and limit
Thickness of infill <5 mm 2 equilibrium models for rigorous analysis of both underground
Thickness of infill >5 mm 3 and open pit projects.
Sheared with no wall contact or thick zones of 4 A benefit of this technique is that rock testing can then be
highly weathered material based on the true rock mass domain variability, rather than
just testing according to rock type. Hence, test samples are
TABLE 8 picked from the domains and major structures and joint sets
identified that are relevant to the engineering structure and its
Joint wall alteration (JWA).
scale. After testing has been completed, the domains can be
Description Logged value RMR rating calibrated against test data
JWA = rock hardness 1 1.00 The value of the domaining process has been realised
in several studies where the presence of major structures,
JWA (dry) weaker than wall 2 0.75 variable rock units or zones of extremely soft rock caused by
rock and filling alteration have been identified as domains.
JWA (wet) weaker than wall 3 0.65 The logging method is unique in that sufficient data is
rock and filling collected to enable the independent determination of all the
major rock mass classification systems such as rock mass
Rock mass ratings can then be used as input to develop rating (RMR), Bieniawski or Laubscher, NGI tunnelling
three dimensional block models using the resource estimation quality index (Q) and geological strength index (GSI).
routines currently available in geological software packages
(Seymour, Dempers and Jenkins, 2007; Jenkins, Dempers The method enables data for a wide range of rock properties
and Seymour, 2009). The domaining methodology allows for and geotechnically significant major structures to be collected
variability of the rock mass to be identified and defined during including rock strength, joint surface condition, fracture
the logging process. Realistic domaining of the geotechnical frequency and fracture orientation.
logging, as opposed to fixed interval logging, enables valid The logging system has been specifically developed to
statistical ranges, averages and quartile values to be readily allow better and more precise appreciation of rock mass and
determined from the block models. A cross-section through structural conditions across the project area thereby optimising
a rock mass model illustrating variability with a single
the use and application of the available geotechnical data
lithological unit is shown in Figure 6.
and improving confidence in the outcomes of geotechnical
The models can also be interrogated based on specific investigations.
geotechnical parameters. The parameters may include rock
strength, shear strength and block size. Matrix codes and
0° - 90° bin joint numbers are also useful for model
REFERENCES
interrogation and refinement, although these parameters are Barton, N, Lien, R and Lunde, J, 1974. Engineering classification of
not used in any of the rating calculations. The ranges of the rock masses for the design of tunnel support, Rock Mechanics,
6:189-236
parameters that are routinely interrogated in the modelling
process are presented in Table 9. An example of the variation Bieniawski, Z T, 1976. A rock mass classification in rock engineering,
in two of these critical parameters within an underground Exploration for Rock Engineering, pp 97-106 (Balkema: Cape
project setting is given in the circled areas of Figures 7 and 8. Town).

SECOND AUSTRALASIAN GROUND CONTROL IN MINING CONFERENCE / SYDNEY, NSW, 23 - 24 NOVEMBER 2010 91