Engineering Geology 199 (2015) 48–61

Contents lists available at ScienceDirect

Engineering Geology

journal homepage: www.elsevier.com/locate/enggeo

The development and application of the alteration strength

index equation
L.D. Wyering a,⁎, M.C. Villeneuve a, I.C. Wallis b, P.A. Siratovich a, B.M. Kennedy a, D.M. Gravley a
a
Department of Geological Sciences, University of Canterbury, P O Box 4800, Christchurch 8140, New Zealand
b
Mighty River Power, 283 Vaughan Road, P O Box 245, Rotorua 3040, New Zealand

a r t i c l e i n f o a b s t r a c t

Article history: We have developed an alteration strength index (ASI) equation to address the effect of hydrothermal alteration
Received 20 April 2015 on mechanical rock properties. This equation can be used to estimate a range of rock strengths, comparable to
Received in revised form 30 July 2015 uniaxial compressive strength (UCS), based on rapid analysis of mineralogy and microstructure. We used rock
Accepted 10 October 2015
samples from three geothermal fields in the Taupo Volcanic Zone (TVZ) to represent a range of alteration
Available online 22 October 2015
types. These are sedimentary, intrusive and extrusive rocks, typical of geothermal systems, from shallow and
Keywords:
deep boreholes (72 measured Depth (mD) to 3280 mD). The parameters used in ASI were selected based on
Empirical relationships literature relating these aspects of mineralogy and microstructure to rock strength. The parameters in ASI define
Uniaxial compressive strength the geological characteristics of the rock, such as proportions of primary and secondary mineralogy, individual
Alteration strength index (ASI) mineral hardness, porosity and fracture number. We calibrated the ASI against measured UCS for our samples from
Porosity the TVZ to produce a strong correlation (R2 of 0.86), and from this correlation we were able to derive an equation to
Microstructural convert ASI to UCS. Because the ASI–UCS relationship is based on an empirical fit, the UCS value that is obtained
from conversion of the ASI includes an error of 7 MPa for the 50th percentile and 25 MPa for the 90th percentile
with a mean error of 11 MPa. A sensitivity analysis showed that the mineralogy parameter is the dominant character-
istic in this equation, and the ASI equation using only mineralogy can be used to provide an estimated UCS range, al-
though the error (or uncertainty) becomes greater. This provides the ability to estimate strength even when either
fracture or porosity information are not available, for example in the case of logging drill cuttings. This research has
also allowed us to provide ranges of rock strengths based solely on the alteration zones, mineralogy, and depth of lithol-
ogies found in a typical geothermal field that can be used to update conceptual models of geothermal fields.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction 2005; Vinciquerra et al., 2005; Smith et al., 2009; Frolova et al., 2010;
Nara et al., 2011; Pola et al., 2012; Heap et al., 2014a; Pola et al., 2014;
Rock strength is necessary for geothermal reservoir development, Wyering et al., 2014; Heap et al., 2015) with reference to how different
management and prospect evaluation because it controls rock behav- rock properties impact the strength of the material.
iour during drilling, stimulation and resource extraction. Tools that Recovering core to test is expensive and, owing to the fractures in
predict rock properties are critical because there are usually limited or the rocks, recovery can be poor leading to only a limited number of sam-
no borehole-based rock property data (Gunsallus and Kulhawy, 1984; ples tested in a given geothermal field. Therefore, many researchers and
Edlmann et al., 1998; Ameen and Smart, 2009). Relationships between industry practitioners apply empirical strength relations to borehole
strength and porosity, density or mineralogy for a specific rock formation geophysics data or limited laboratory data (Edlmann et al., 1998;
have been widely developed based on laboratory tests on rock core from Koncagül and Santi, 1999; Dinçer et al., 2004; Entwisle et al., 2005;
a given field or lithology (Chang et al., 2006; Tamrakar et al., 2007; Çobanoğlu and Çelik, 2008; Binal, 2009). Chang et al. (2006) reviewed
Rigopoulos et al., 2010; Singh et al., 2012; Karakul and Ulusay, 2013). thirty-two empirical relationships for sedimentary rocks where physical
These relationships, however, were developed using mainly sedimenta- rock properties were derived from borehole geophysics. Their review
ry, granitic and metamorphic rock samples and cannot be applied ubiq- made clear that a few of the empirical relationships appeared to work
uitously to all lithologies, especially hydrothermally altered volcanic fairly well for some subsets of the rocks studied. Wyering et al., 2012
rocks. Only recently have studies investigated the physical and mechan- assessed the applicability of selected empirical equations for predicting
ical properties of volcanic rocks (Ladygin et al., 2000; Frolova et al., uniaxial compressive strength (UCS) for geothermally altered litholo-
gies and found that the correlations between predicted UCS and mea-
⁎ Corresponding author at: Department of Geological Sciences, University of
sured UCS were poor. The downfall of these empirical relationships is
Canterbury, Private Bag 4800, Christchurch 8140, New Zealand. that they are only applicable to the particular lithologies being studied,
E-mail address: [email protected] (L.D. Wyering). and do not necessarily correlate for all rock types, especially silicic

http://dx.doi.org/10.1016/j.enggeo.2015.10.003
0013-7952/© 2015 Elsevier B.V. All rights reserved.
L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61 49

volcanic rocks affected by secondary mineralisation. Whilst the equa- structural boundaries, volcanic vent positions and geothermal systems
tions presented in Chang et al. (2006) may be useful to a practitioner (Fig. 1: Cole, 1990; Wilson et al., 1995). The N20 geothermal systems
in the geothermal industry as a first order approximation, they are in the TVZ, totaling ~ 4500 MW thermal output (Bibby et al., 1995),
focused on sedimentary rocks with no high-temperature secondary are related to magmatic heat generated at depth and shallow crustal
mineralisation and therefore have limited utility (Yagiz, 2009). structure that provides the permeability necessary for convective trans-
Research has shown that several rock properties (mineral hardness, port of hydrothermal fluids (Rowland and Sibson, 2004; Rowland and
secondary minerals, microstructural damage that includes the presence Simmons, 2012). These circulating fluids become rich in dissolved min-
of microfractures and pores) can influence the predicted rock strength erals, as they percolate through the stratigraphy (Henneberger and
of material (Tuğrul and Zarif, 1999; Ameen and Smart, 2009; Browne, 1988) and precipitate minerals in the reservoir rocks produc-
Rigopoulos et al., 2010; Coggan et al., 2013; Heap et al., 2014a). Several ing the secondary mineralisation that are observed when the rocks are
petrographic and weathering indices related to chemical, petrological drilled and brought to the surface (Goff and Janik, 2000). The rock
and mechanical properties, have been suggested to identify the impact types we used in this study (described in detail in Wyering et al.,
of alteration on rock properties in different lithologies (Ulusay et al., 2014) were sourced from shallow formations – Rhyolitic ignimbrite,
1994; Tamrakar et al., 2007; Ceryan et al., 2008; Yildiz et al., 2010; Rhyolitic lava, and Siltstone/Sandstone – and from deep formations –
Pola et al., 2012, 2014). Rhyolitic ignimbrite, Andesite Lava/Breccia and Tonalite intrusive –
This paper describes the development of a strength prediction equa- from numerous geothermal fields in the TVZ.
tion that can be used to calculate a strength range comparable to UCS
using descriptions of hydrothermal alteration, secondary mineralisation,
porosity and bulk rock structural damage. The core samples used are 3. Data source
sourced from the Ngatamariki, Rotokawa and Kawerau geothermal
fields from the Taupo Volcanic Zone (TVZ), New Zealand, allowing the All of the data used in this study are sourced from Wyering et al.
equation to be adapted for geothermal fields located in the TVZ. It (2014). They characterized the physical and mechanical properties of
encompasses a variety of lithologies that are found in the TVZ and the lithologies from the Ngatamariki, Rotokawa and Kawerau geothermal
differing geothermal environments they are exposed to. The equation fields (Fig. 1), using non-destructive and destructive methods to deter-
could be used in other geothermal systems worldwide with similar geo- mine porosity, density, ultrasonic wave velocities and uniaxial com-
thermal conditions or adapted easily to suit. We will show that develop- pressive strength (UCS). The samples were cored to a mean diameter
ment of this equation has improved understanding of how alteration of 39.6 mm and were cut and ground to within the length to diame-
mineralogy and physical properties control rock strength. We will ter ratio of 2:1. Their study examined thin sections using a polarized
demonstrate how a variant of the equation could be used in the field light microscope, that utilized plane polarized light (PPL) and cross-
to optimize drilling of geothermal reservoirs through improved drill polarized light (CPL) to identify primary and secondary minerals (that
bit selection. includes but is not limited to clays, quartz, epidote, chlorite, albite and
pyrite), microfractures and bulk rock fractures in the lithologies.
2. Geothermal setting Although Wyering et al. (2014) did mention the textures of the sam-
ples, they were not used in this study because the samples were
The active Taupo Volcanic Zone (TVZ) is located at the southern end moderately to intensely altered. The textures within the samples
of the Tonga Kermadec arc in the central North Island of New Zealand, in were completely replaced and difficult to distinguish, reducing the
a 300 km long (200 km on land) and 60 km wide belt, defined by caldera ability to use the data.

Fig. 1. A map of geothermal activity in the Taupo Volcanic Zone (TVZ), showing the positions of geothermal systems, the active and inferred caldera boundaries and the Taupo Rift
(white lines with arrows). The geothermal fields used in this study are located. Abbreviations are named calderas: KA = Kapenga, MO = Mangakino, OH = Ohakuri, OK = Okataina,
RE = Reporoa, RO = Rotorua, TA = Taupo, WH = Whakamaru. The map is split up into the main volcanic activity in the TVZ and outlined by the boundary of the young TVZ
(b0.34 Ma) (Adapted from Wilson et al., 1995; Bibby et al., 1995; Rowland and Sibson, 2004; Kissling and Weir, 2005; Rowland and Simmons, 2012).
50 L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61

Table 1 Table 3
The semi-quantitative categories for the primary minerals and the percent- Hardness index values assigned to Moh's hardness.
ages representing the categories.
Moh's hardness scale Hardness index
Category Representative percentage
N7 2.1
Abundant (A) 50 7–6 1.7
Common (C) 25 6–4.5 1.3
Minor (M) 10 4.5–3 0.9
Rare (R) 5 3–2 0.5
b2 0.1

For this study we used their connected porosity, (Ф herein referred relative percentage alteration. This portion of the equation, the
to as porosity), UCS results, and thin section analysis of the lithologies. mean mineralogy parameter, accounts for the overall impact of all
Wyering et al. (2014) used the suggested method from the International mineral types on the strength.
Society of Rock Mechanics (ISRM) (Ulusay and Hudson, 2007a) to Φ is the connected porosity, as a volume %, which is related to both
determine the porosity (as volume %) and density of the hydrother- original porosity and changes in porosity due to mass exchange during
mally altered samples using cylindrical cores. They did not measure hydrothermal alteration processes. It is multiplied to the mineralogy
total porosity (sum of connected and closed porosity) because its mea- portion of the equation in order to have a power law-based impact on
surement requires grinding the sample into a powder, which is incom- the strength estimate, where the higher the porosity, the lower the
patible with further tests, such as UCS. UCS testing was completed using strength.
the ISRM suggested methods (Ulusay and Hudson, 2007b) and the To represent the presence of fractures, the fracture index, Snf a num-
American Society for Testing and Materials (ATSM, 2010). The samples ber between 0 (no fractures) and 6 (multiple large open fractures), is
were tested using a Technotest 3000 kN, servo-controlled loading frame assigned to each sample. This parameter is multiplied to the mineralogy
and loaded at a constant stress rate (between 0.02 kN/s to 0.500 kN/s) and porosity portions of the equation to reduce the strength index value
to ensure failure occurred within 5–10 min. Tokyo Sokki Kenkyujo based on the severity of fracturing.
Co. Ltd. (TML) 20 mm strain gauges with a factor of 2.12 were glued
to the samples; two axial and two radial. The samples were tested at
ambient laboratory temperature and humidity conditions. 4.1. Alteration strength index (ASI) development

4. Proposed equation — alteration strength index (ASI) 4.1.1. Mineralogy (Pm and Sm) and alteration index (AI)
Primary and secondary mineralogy both influence the strength of
The alteration strength index equation (ASI), developed in this rocks, where rocks with predominately weak minerals lead to a
study, is given in Eq. (1). In this section, we introduce the equation lower strength compared to rocks containing predominantly strong
and explain the meaning of each of the parameters, and how they relate minerals (Vutukuri et al., 1974; Rigopoulos et al., 2010; Li et al.,
to each other. The purpose of the equation is to estimate of rock strength 2012; Villeneuve et al., 2012). We developed a method for deriving
based on physical and mineralogical properties, which can be undertak- a representative percentage of the primary and secondary minerals to
en rapidly in the field. This method does not require establishing the pri- address this. Identifying the exact percentages of minerals in hydrother-
mary lithology of the material to estimate rock strength, but rather mally altered samples can be difficult (Stringham, 1952), and identify-
focuses on the physical characteristics of the rocks as they are found. ing alteration mineralogy, including clay, requires complex separation
We selected the parameters that make up the equation were based on and thorough XRD analysis (Hillier, 2000), which are not suitable for
published research that shows which key parameters affect the strength rapid, field application. Consequently, the mineral analysis in our
of both hydrothermally altered and unaltered rocks. study was completed using modal percentage estimates from thin
sections.

S
ASI ¼ ðPm ð1−AIÞ þ Sm AIÞ Ф −0:03 1− nf ð1Þ To determine the modal percentages of the primary and second-
25 ary minerals the primary and secondary minerals need to be identi-
fied. The identified minerals are then assigned representative modal
Where, Pm (primary mineralogy, i.e. minerals associated with the percentages. The primary mineralogy is accounted for by assigning
original lithology) and Sm (secondary mineralogy, i.e. minerals asso- a semi-quantitative category: abundant (A), common (C), minor
ciated with alteration processes) are representative values based on (M) and rare (R), related to abundance in thin section. A representative
the hardness of each mineral present and the relative proportion of
each mineral contained in each sample (obtained from thin section
or visual examination of samples or drill cuttings). Alteration index
(AI) is the percentage of the sample that has changed from the original Table 4
An example of how to determine the primary and secondary values of the mineralogy
due to hydrothermal alteration (primary mineral replacement, infilling
parameter that is multiplied by the alteration index (AI). The assigned percentage is mul-
of fractures and voids), and acts to scale P m and Sm according to tiplied by the hardness index and added together to produce the resulting Pm and Sm
values.

Category Mineral Assigned percentage Hardness index

Table 2 Primary mineral (Pm)
The semi-quantitative categories for the secondary minerals and the percentages A Plagioclase 25 1.7
representing the categories. M Quartz 10 2.1
Pm = 35.7
Ranking 1 2 3 4 5 6 7 8 9 10 11
Secondary mineral (Sm)
Representative percentages of secondary minerals
A Quartz 50 2.1
Abundant (A) 50 C Calcite 40 0.9
Common (C) 45 40 35 30 25 C Epidote 35 1.7
Minor (M) 20 17.5 15 12.5 10 7.5 5 R Chlorite 4.5 0.5
Rare (R) 5 4.5 4 3.5 3 2.5 2 1.5 1 Sm = 202.75
L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61 51

Fig. 2. Relationship between uniaxial compressive strength (UCS) and porosity for the Rotokawa andesite samples.

percentage of 50, 25, 10, 5, respectively, is assigned to each semi- 4.2. Porosity (Φ)
quantitative category (Table 1).
This is repeated for the secondary minerals; however, due to the Igneous rocks can be intrusive – made primarily of interlocking crys-
abundance and variety of the secondary minerals typically found in tals, extrusive – composed of phenocrysts and groundmass (which may
hydrothermally altered rocks, the semi-quantitative categories are be microcrystalline or contain glass), or pyroclastic – composed of
additionally ranked between 1 and the total number of secondary grains of crystals, glass, and rock that are cemented or welded together.
minerals within each category (Table 2). This allows for samples with Porosity in rocks is fundamentally controlled by the efficiency and
multiple minerals that are abundant, common, minor or rare. Table 2 extent of outgassing of the magma. The porosity is present as bubbles
illustrates the representative percentages of the secondary minerals (in magma) and can be preserved as vesicles (in volcanic rock) (Dobson
based on this ranking. If a sample has chlorite as a common (C) mineral, et al., 2003; Shea et al., 2010; Heap et al., 2014b). As alteration occurs the
but it is the 3rd most common mineral, it would be assigned a represen- porosity within a sample can increase or decrease through mass trans-
tative value of 35. fer with dissolution and precipitation of minerals as the fluids permeate
Vutukuri et al. (1974) reported that there is a positive relationship through the rocks, thereby creating or infilling voids (Ferry, 1979;
between the hardness of minerals present in the rock and rock strength. Giggenbach, 1984; Henneberger and Browne, 1988; Reyes, 1990;
We, therefore, used hardness as a proxy for mineral strength to estimate Simmons and Browne, 2000; Esmaeily et al., 2012). The pore structure
the aggregate strength of a sample. To determine whether a mineral is is dependent on numerous factors such that the porosity within a rock
to be classified as soft or hard in the ASI equation, we took minerals can vary greatly (Hudyma et al., 2004; Heap et al., 2014b). Porosity is
with Moh's hardness less than 5 to be soft and minerals with Moh's an important factor in rock strength, because voids reduce the integrity
hardness greater than 5 to be hard, as described in Broz et al. (2006) of the material and, as indicated by the ISRM (Ulusay and Hudson,
and Whitney et al. (2007). The minerals were assigned hardness index 2007a), even a small volume of pores can have a noticeable mechanical
values from 0.1–2.1. Soft minerals were assigned hardness index value effect (Sammis and Ashby, 1986; Fakhimi and Gharahbagh, 2011; Heap
less than 1 (0.1–0.9) because their presence tends to ‘weaken’ the et al., 2014a, 2014b). Pores reduce stiffness and strength due to stress
rock, while hard minerals were assigned hardness index value greater concentration on the boundary of the pores. Additionally, the pores
than 1 (1.3–2.1) because their presence tends to ‘strengthen’ the rock
(Table 3).
The hardness index parameter for a mineral in a sample is multiplied
Table 5
by its representative percentage value to obtain Pm and Sm for primary
Values assigned to the samples based on the fractures seen in thin section and on the bulk
and secondary minerals, respectively (See Table 4 for a worked exam- rocks.
ple). The result is that an abundant mineral has a greater influence on
Fracture Fracture type Fracture size
the whole rock strength estimate than a mineral with low abundance.
value (Snf)
The approximate proportion of primary versus secondary mineralogy
is estimated, and this produces the alteration index (AI), e.g. a value of 0 No fractures in sample –
1 Microfractures seen in Narrow fractures in thin section
50 represents a sample with half primary and half secondary minerals. thin section
The summed mineralogical influences for primary and secondary min- 2 Small fracture — closed 1 fracture b1 mm in width, b10 mm in
erals are then weighted using the AI value to conclude the mean miner- length and fracture remains closed
alogy parameter portion of the ASI equation. 3 Small fracture — open 1 fracture b1 mm in width, b10 mm in
length and fracture remains opened
Future studies could make use of Scanning Electron Microscope
4 Large fracture — closed 1 fracture N1 mm in width, N10 mm in
(SEM) in the form of Quemscan or robust analysis of Energy Dispersive length and fracture remains closed
Spectrometry (EDS) maps and spectra alongside optical petrology to 5 Large fracture — open 1 fracture N1 mm in width, N10 mm in
assist in quantifying the mineralogy percentages in a research environ- length and fracture remains open
ment; however as mentioned previously this is unlikely to be feasible to 6 Multiple large fractures — 2 or more fractures N1 mm in width,
closed/open N10 mm in length
carry out in-field for real-time analysis.
52 L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61

Fig. 3. Relationship between uniaxial compressive strength (UCS) and fracture values for the Ngatamariki andesite samples.

may be filled with fluids, which may help in crack propagation under Studies investigating the impact of controlled microstructural dam-
compression due to the incompressibility of the fluids, leading to an in- age in rocks found that peak strength is sensitive to the amount of in-
crease in stress concentrations at pore boundaries as the fluid is trying duced damage, resulting in lower peak strengths (Martin, 1997; Heap
to escape (Price, 1960; Lama and Vutukuri, 1978; Luping, 1986; Baud et al., 2015). Wyering et al., 2014 show the same results for the andesite
et al., 2000, 2009; Brantut et al., 2013). breccia from Ngatamariki, where 11 out of 29 samples had pre-existing
The measurement of porosity likely includes the presence of pre- fractures. The samples that contained pre-existing fractures had a mean
existing, open fractures, because the saturation fluid is able to infiltrate UCS of 32.4 ± 7.3 MPa and an average porosity of 2.2% (samples NM7-5,
these fractures during saturation. In our equation, the fracture index 7-8, 7-12), while the remaining samples had a mean strength of
(Snf) accounts explicitly for the high potential for failure along pre- 117.5 ± 45.9 MPa and average porosity of 1.6%. To account for this we
existing, but closed or incipient, fractures, that have less proportional in- selected index values for Snf between 0 and 6 for seven different catego-
fluence on the measurement of porosity than pores, due to their smaller ries of visible macrofractures, in order of severity. Consequently, a large
void volume (Mueller et al., 2005). In Wyering et al., 2014, samples with fracture through the bulk sample is assigned a higher value than a sam-
open fractures tended to break during sample preparation and thus ple that has evidence of many thin short microfractures in thin section
tended to be excluded from the data set. (Table 5).
To isolate the impact of porosity on UCS we used the Rotokawa an- To isolate the impact of pre-existing fractures on UCS we used a sub-
desite samples (Siratovich et al., 2014; Wyering et al., 2014), which set of the Ngatamariki andesite breccia samples that had similar prima-
have similar mineralogy and fracture characteristics (fracture index ry and secondary mineralogy and porosity (ranging from 1.4–2.3%). The
values of 0–1) with each other. We found that gaps in the data made relationship between UCS and fractures is linear and negative (Fig. 3),
the decision between power, log or linear trends between the porosity and we derived the formulation (1 − Snf/25) used in the ASI equation
and UCS in our data difficult. We selected an inverse power relationship from this fit. This formulation results in no impact if there are no frac-
because many authors have reported an inverse power correlation for a tures (Snf = 0), and greater strength reduction as the severity of frac-
variety of rock types (Lama and Vutukuri, 1978; Palchik, 1999; Begonha tures increases (Snf = 1 to 6). The constant 1/25 scales the 0–6 range
and Sequeira Braga, 2002; Li and Aubertin, 2003; Sousa et al., 2005; for Snf based on the results from the andesite breccia.
Chang et al., 2006; Heap et al., 2014b). We have used the resulting in- To allow for the use of the fracture parameter if only drill cutting are
verse power correlation (Fig. 2), with the power constant − 0.13, in available, the veining abundance could be used as a substitute. Fractures
the ASI equation based on this analysis. in geological systems are important conduits for fluid flow (Wangen
and Munz, 2004; Rawal and Ghassemi, 2010). Ion rich hydrothermal
fluids travel through fractures, textures and faults depositing secondary
4.3. Fracture index (Snf) minerals creating veins. We assume that veining corresponds to frac-
tures to use veining as a basis for fracture estimating. The veining abun-
Some of the core used in the Wyering et al., 2014 study contained dance substitute values are displayed in Table 6. We have used fractures
pre-existing fractures and veins, making preparation difficult. There from core to produce the values in this study; however, this can be dif-
are three types of fractures possible in a rock: intergranular — occur ficult if the drill cuttings are small.
between grain, intragranular — occur within a grain, and transgranular —
affect more than one grain (Sousa et al., 2005). The majority of the pre-
Table 6
existing fractures in the samples used for this study are transgranular.
Fracture parameter substitute for the drill cuttings.
Micro- and macrofractures, whether pre-existing or induced during
testing, coalesce during uniaxial compression ultimately leading to fail- Veining abundance Snf
ure of the sample (Bieniawski, 1967; Bieniawski et al., 1969). Samples No veins in samples (R) 0
that contain pre-existing fractures require less energy to propagate Micro veins (M) 1
the fractures, resulting in lower peak strength values in these samples Small veins in the cuttings b1 mm (C) 2
Large veins in the cuttings N1 mm (A) 4
(Walsh, 1961; Martin, 1997; Siratovich et al., 2014).
L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61 53

Fig. 4. Relationship between alteration strength index (ASI) and uniaxial compressive strength (UCS) showing a power relationship.

5. Discussion allow the ASI equation to provide ranges of strength for different for-
mations and lithologies based on different rock conditions/properties,
The ASI equation has been developed as a tool to rapidly estimate including pre-existing fractures. Values used to calculate ASI from the
the range of strengths for a hydrothermally altered rock based on its Ngatamariki, Rotokawa and Kawerau Geothermal fields are in Appen-
present geological characteristics using data collection techniques cur- dix A Table A.1. The correlation of ASI to measured UCS (Fig. 4) shows
rently used in field mapping and drilling exploration. In principle, orig- a clear trend (r2 of 0.86) — such that the ASI can be used to predict
inal lithology and texture will likely affect rock strength, but the process the UCS rock strength. To convert the ASI value to UCS Eq. (2) can be
for interpreting original lithology and texture do not lend themselves to used. Our dataset does not include samples with ASI values below 60,
rapid strength estimation. We have devised the ASI to use rock charac- however rocks with such low ASI could be encountered, for example
teristics that are easily observed and quantified in the field, irrespective if samples have extremely high porosity. According to Fig. 4 and
of its original lithology, and to relate them to laboratory strength values, Eq. (2), however, the strength would be so low as to barely be classified
such as UCS. as rocks. The UCS results from both laboratory tests and estimated from
Wyering et al., 2014 prepared the samples used to devise this ASI are in Appendix A, Table A.2.
equation for UCS testing according to internationally recognized stan-
dards (ISRM/ASTM). The presence of fractures influences the tested UCS ¼ 7 10−8 ASI 4:3661 ð2Þ
rock strength (Fig. 3), which, according to ISRM and ASTM standards
would result in “invalid” tests. These fractures will impact drilling per- The empirical fit on which Eq. (2) is based contains variability and
formance, thus the need to include fractures to derive strength esti- the UCS calculated from the ASI should always be quoted with a range
mates applicable to drilling. We have devised the Snf parameter to of error. This will allow for the range to cover natural variations in the

Fig. 5. Relationship between the calculated UCS derived from ASI and measured uniaxial compressive strength (UCS) with the absolute error ranges — 50th (dark grey lines) and 90th
percentile (light grey lines).
54 L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61

Fig. 6. Relationship between uniaxial compressive strength (UCS) and the mineralogy parameter from the three geothermal fields showing a power relationship. The change in colour
represents the different geothermal fields (Kawerau is light grey, Ngatamariki is black and Rotokawa is dark grey). The different shapes represent the different lithologies — andesite
(square), ignimbrite (circles), intrusive (crosses), sedimentary (diamonds) and rhyolite (triangles).

rock and error arising from visual examinations of mineral abundance. intensity, with calculated mean mineralogy parameter) and measured
The 50th and 90th percentile absolute error ranges for this dataset are UCS are shown in Table 7. The order of abundance of the main minerals
± 7 MPa and ± 25 MPa respectively, with a mean error of ± 11 MPa. namely quartz and chlorite, demonstrates how the abundance and
This error value could be used to represent this variability. Fig. 5 hardness of minerals would cause an increase or decrease in the mean
shows a plot of calculated UCS from ASI against measured UCS from mineralogical parameter. The comparison to UCS shows that the mean
testing and also shows the position of these 50th and 90th percentile mineralogy parameter alone indicates that the Rotokawa andesite is
error ranges. stronger than the Kawerau andesite.
The mean mineralogy parameter with either the fracture parameter
5.1. Modified alteration strength index (mASI) or the porosity parameter versus measured UCS (Figs. 7 and 8, respec-
tively) show weaker correlations than the complete ASI correlation
To explore the potential for a modified ASI, based on fewer input (Fig. 3). The deep ignimbrites from Ngatamariki have clusters with sim-
parameters, we plotted each parameter in the ASI equation against ilar mineralogy (black circles, Fig. 6), however when the fracture param-
UCS to examine its overall influence on the result (Figs. 6–8). Fig. 6 illus- eter, which spans the full range (0–6) (Fig. 7) or the porosities (3%–20%)
trates the strong relationship observed between the mineralogy param- (Fig. 8) are taken into account, the points no longer cluster and are well
eter and the measured UCS. spread along the data trendline.
The mineralogy aspects of the Rotokawa andesite and Kawerau These observations show that the mean mineralogy parameter is the
andesite (order of the top three secondary minerals and alteration dominant characteristic in the ASI equation; however, the inclusion of

Fig. 7. Relationship between uniaxial compressive strength (UCS) and the mineralogy and fracture parameters from the three geothermal fields showing a power relationship. The change
in colour represents the different geothermal fields (Kawerau is light grey, Ngatamariki is black and Rotokawa is dark grey). The different shapes represent the different lithologies —
andesite (square), ignimbrite (circles), intrusive (crosses), sedimentary (diamonds) and rhyolite (triangles).
L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61 55

Fig. 8. Relationship between uniaxial compressive strength (UCS) and the mineralogy and porosity parameters combined from the three geothermal fields showing a power relationship.
The change in colour represents the different geothermal fields (Kawerau is light grey, Ngatamariki is black and Rotokawa is dark grey). The different shapes represent the different
lithologies — andesite (square), ignimbrite (circles), intrusive (crosses), sedimentary (diamonds) and rhyolite (triangles).

the effects of porosity and fractures leads to a stronger relationship with calibrated with hydrothermally altered rocks from three different geo-
UCS (Fig. 4). The best way to use ASI is in its original form, however if thermal fields with diverse primary lithologies from both shallow and
porosity or pre-existing fracture or veining data are unavailable, good deep alteration zones typical of the TVZ and should be suitable for any
predictions of UCS are still possible. For example when using drill hydrothermally altered rocks in this area. The samples tested represent
cuttings, if porosity and fracture data are not available the mASI a majority of lithologies found in a variety of geothermal fields that are
equation using the mean mineralogy parameter alone can be used (r2 exposed to differing hydrothermal environments. Because the ASI
of 0.79) (Fig. 6). The 50th and 90th percentile absolute error ranges provides ranges of rock strengths it can be used to provide a first-pass
for this dataset are ± 9 MPa and ±44 MPa respectively, with a mean estimate for strength for hydrothermally altered rocks in similar envi-
error of ± 18 MPa. The fracture parameter shows a much greater ronments, although further testing is necessary to determine if the con-
improvement of the fit of the mASI (r2 of 0.85) (Fig. 7) than the porosity stants require calibration. Therefore, should this equation be used in
data (r2 of 0.79) (Fig. 8). The 50th and 90th percentile absolute error active or non-active geothermal environments with different lithologies
ranges for this dataset are ±8 MPa and ±24 MPa respectively, with a than those tested in this study, a user could tests the ASI vs. UCS (Fig. 4)
mean error of ± 11 MPa. Although the difference between mASI with relationship for the new lithologies and modify the equation to suit the
only the mean mineralogy parameter and mASI with mean mineralogy new environment. If this is not possible, it should prove a good first-pass
parameter and porosity is negligible in terms of r2, it is worth noting for estimating rock strength. Any additional information added to the
that the presence of the porosity parameter in ASI leads to a slight results will only but improve the reliability of the equation.
improvement in the correlation between measured UCS and ASI (r2 of This research has also allowed us to provide ranges of rock strengths
0.86) (Fig. 4) over mASI with mean mineralogy parameter and fracture based solely on the alteration zones, mineralogy, and depth of litholo-
parameter only. When using the mASI it is necessary to be aware of the gies found in a typical geothermal field that can be used to update con-
reduction in accuracy of the prediction of UCS, as reflected in the wider ceptual models of geothermal fields (Fig. 9). This does not include rock
error ranges. The mASI could be used to give a first estimate of the rock softening due to increased temperature from the geothermal gradient
strength from drill cuttings. When possible thin section can be make of (Karfakis, 1985; Kusznir and Park, 1987; Weinberg and Podladchikov,
the cuttings and analysed in further detail on site with the correct 1994) or compactant modes of failure for porous rock at depth (Heap
equipment. The additional effort will provide additional confidence in et al., 2015). This conceptual model adds an important element to
the results through better mineralogy and fracture/veining abundance understanding how rock mechanics plays a part in a geothermal system,
observations. and can be a fundamental addition to the geothermal industry to sup-
port drilling, wellbore stability studies and mechanical modelling.

5.2. Predictive capabilities

6. Conclusions
It is important to understand how hydrothermal alteration has influ-
enced the rock properties of a lithology in order to make strength-based 1. We developed the alteration strength index (ASI) equation to predict
decisions for engineering works in geothermal settings. For example, rock strength based on different geological characteristics, including
predicting strength changes can improve drilling efficiency and help mineralogy, fractures and porosity for hydrothermally altered rocks.
guide drill bit selection. The ASI equation can be used in the lab or in It has been developed so that only the mineralogy, porosity and frac-
the field to provide a range of strengths. The ASI equation has been ture state are needed to rapidly determine a range of rock strengths

Table 7
The top three minerals, alteration intensity, mean mineralogy parameter and UCS (MPa) of the andesite lithology from Rotokawa and Kawerau.

Lithology Top three minerals Alteration intensity Mean mineralogy parameter Mean UCS

Rotokawa Andesite Calcite, quartz and chlorite 70–95% 139.1 127.8

Kawerau Andesite Calcite, chlorite and quartz 95% 118.13 67.2
56 L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61

mean error of ±11 MPa. These can be used to define ranges of esti-
mated rock strength.
4. The mineralogy is the dominant characteristic in this equation, and
forms one possible modified ASI (mASI) formulation, however a larg-
er error range is specified for mASI. The inclusion of pre-existing frac-
tures in addition to mineralogy makes the relationship of mASI to
UCS stronger. The inclusion of porosity data only shows improve-
ment in correlation to UCS for the ASI is in its original form. This
makes our approach functional in a field setting to provide real-
time strength estimates during engineering works, such as drilling
or excavation, where porosity and pre-existing fractures can be diffi-
cult to assess.
5. The conceptual model of geothermal systems with alteration zones,
temperature profiles and fluid path migration with associated rock
strength profiles based on the ASI shown in Fig. 9 adds an important
element to understanding how rock mechanics plays a part in a geo-
thermal system. This can form the basis for constructing site-specific
(including actual stratigraphy, unit and temperature depths, and
structural data) conceptual models, which is a valuable addition to
the information needed by the geothermal industry to support dril-
ling optimization, wellbore stability studies and mechanical, struc-
tural and hydrogeological modelling.

Conflict of interests
Fig. 9. Conceptual model of a conventional, hot, liquid dominated geothermal field. The
model has been split into the alteration zones typical for a geothermal field, with temper-
ature profiles, surface expressions with the addition of strength profiles (adapted from The authors declare that they have no competing interest.
Cumming, 2009).

Acknowledgements
for any hydrothermally altered rocks, whether in active or non-active
geothermal systems. The authors wish to thank Mighty River Power, Rotokawa Joint Ven-
2. To represent a variety of alteration zones and resulting alteration ture Limited; a joint venture between the Tauhara North No.2 trust and
mineralogy, a range of lithologies were tested from core sampled Mighty River Power Limited, and Ngati Tuwharetoa Geothermal Assets
from the Ngatamariki, Rotokawa and Kawerau geothermal fields in Limited, for the use of core supplied for this study. The staff of the De-
the Taupo Volcanic Zone. These rocks include shallow and deep for- partment of Geological Sciences at the University of Canterbury was in-
mations of extrusive and intrusive rocks that contained large quanti- valuable in assisting in all aspects of this research. The Brian Mason
ties of primary and secondary minerals including but not limited to Trust also provided for funding for the collection and transportation of
clays, quartz, calcite, chlorite, albite, pyrite and epidote. the core to the University of Canterbury. The Callaghan Innovation (con-
3. Our results show a strong relationship between ASI and measured tract number: MRPR1201/32965) and Source to Surface programme, a
UCS (R2 of 0.86), and were used to derive an equation to convert multiyear research initiative between the University of Canterbury
ASI to UCS. The 50th and 90th percentile absolute error ranges of and Mighty River Power who provided funding for the completion of
this relationship are ± 7 MPa and ± 25 MPa respectively, with a fieldwork and core collection.

Appendix A

Table A.1
Values of the parameters used in the ASI equation for the samples from the Ngatamariki, Rotokawa and Kawerau Geothermal fields (Wyering et al., 2014). AI = alteration index, Pm =
primary mineralogy, Sm = secondary mineralogy, Snf = fracture factor, Ф = porosity.

Sample ID AI Pm Sm Snf Ф

Shallow formations G1 Box 13 1-1 0.85 186.0 79.4 0 25.64

Rhyolitic ignimbrite G1 Box 13 1-2 0.85 186.0 79.4 0 25.14
G1 Box 13 1-3 0.85 186.0 79.4 0 24.75
G1 Box 13 2-1 0.85 183.3 79.4 0 25.51
G1 Box 13 2-2 0.85 186.0 79.4 0 24.79
G1 Box 13 2-3 0.85 186.0 79.4 0 25.40
G1 Box 13 2-4 0.85 186.0 79.4 0 26.51
G1 Box 11 1-1 0.85 183.3 110.5 0 19.80
G1 Box 11 1-2 0.85 183.3 110.5 0 17.58
G1 Box 11 1-3 0.85 183.3 110.5 0 18.37
G1 Box 11 1-4 0.85 183.3 110.5 0 18.79
G1 Box 11 2 0.85 192.2 110.5 0 18.15
G1 Box 17 1 0.8 196.7 41.4 0 46.32
G1 Box 17 2-1 0.8 196.7 44.8 0 46.06
G1 Box 17 2-2 0.8 196.7 41.4 0 47.48
G1 Box 17 3 0.8 196.7 41.4 0 46.73
G1 Box 17 4 0.8 196.7 41.4 0 48.03
L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61 57

Table A.1 (continued)

Sample ID AI Pm Sm Snf Ф
Shallow formations G2 Box 15 1 0.85 190.0 78.6 0 25.33
Rhyolitic ignimbrite G2 Box 15 2-1 0.85 190.0 78.6 0 25.28
G2 Box 15 2-2 0.85 190.0 78.6 0 24.97
G2 Box 15 3 0.85 190.0 78.6 0 26.29
G2 Box 15 4 0.85 190.0 78.6 0 24.88
G2 1-1 0.95 210.0 84.1 0 54.66
G2 1-2 0.95 210.0 84.4 0 56.56
G2 1-3 0.95 196.7 84.4 0 47.00
G2 2-1 0.95 196.7 84.1 0 50.15
G2 2-2 0.95 210.0 84.1 0 47.91
Rhyolitic lava KAM 11 0.95 190.0 73.3 3 19.54
Siltstone/Sandstone KA30 1-1 0.8 206.4 90.2 1 17.40
KA30 1-2 0.8 203.3 90.2 1 16.99
KA30 2-1 0.8 206.4 90.2 0 16.10
KA30 2-2 0.8 206.4 90.2 0 16.10
KA30 3-1 0.8 203.3 90.2 1 17.43
KA30 3-2 0.8 203.3 90.2 0 17.27
KA30 4-1 0.8 203.3 90.2 0 18.82
KA30 4-2 0.8 203.3 90.2 1 18.87
KA30 4-3 0.8 203.3 90.2 1 19.44
KA30 4-4 0.8 203.3 90.2 1 19.86
KAW 17 1 0.9 190.0 87.8 1 14.95
KAW17 3-2 0.8 190.0 91.3 1 17.47
KAW 17 C2 4 1 0.0 72.3 0 23.03
Deep formations NM1 1-2 0.9 175.0 126.2 2 6.17
Rhyolitic ignimbrite NM1 1-1 0.9 175.0 126.2 0 4.70
NM2 2-3 0.95 170.0 98.8 2 20.28
NM2 2-1 0.95 170.0 98.8 1 18.60
NM2 2-2 0.95 170.0 95.0 1 20.07
NM 3 3-2 0.95 170.0 100.9 4 9.00
NM3 3-1 0.95 170.0 100.9 3 9.52
NM3 3-3 0.95 170.0 99.8 4 10.90
NM3 3-7 0.95 170.0 100.9 3 9.10
NM3 3-4 0.95 170.0 99.8 4 9.79
NM 4 0.98 210.0 117.8 3 5.44
NM 4 0.98 210.0 124.3 6 6.17
NM 5 5-2 0.85 178.0 110.4 2 12.10
NM 5 5-1 0.85 178.0 110.4 2 9.90
NM8A C1-1 0.95 180.0 118.2 2 4.03
NM8A C1-2 0.95 180.0 118.2 2 3.17
NM8A C1-3 0.95 180.0 123.7 2 3.57
NM8A C1-5 0.95 180.0 123.7 0 3.58
NM8A C1-6 0.95 180.0 118.2 2 4.09
NM11 1-2 0.95 180.0 112.1 3 15.29
NM11 1-3 0.95 180.0 112.1 3 15.74
NM11 1-4 0.95 180.0 112.1 3 14.78
NM11 1-5 0.95 180.0 112.1 3 14.65
KA37 1-1 0.9 190.0 102.2 0 17.21
KA37 1-2 0.9 190.0 102.2 0 17.60
KA37 1-3 0.9 190.0 106.1 0 17.53
KA37 1-4 0.9 190.0 106.1 0 17.68
KA37 1-5 0.9 190.0 102.2 2 17.49
KA37 1-6 0.9 190.0 102.2 0 17.72
KA37 1-7 0.9 190.0 102.2 0 17.79
KA37 1-8 0.9 190.0 102.2 0 16.99
KA37 2-1 0.9 190.0 106.1 0 18.73
KA37 2-2 0.9 198.6 106.1 0 17.86
Andesite NM 7-1 0.9 186.0 148.2 1 1.50
Lava/breccia NM 7-2 0.95 180.0 147.1 0 1.69
NM 7-3 0.9 186.0 148.2 1 1.63
NM 7-4 0.9 186.0 148.2 2 1.63
NM 7-5 0.95 180.0 147.1 5 2.27
NM 7-6 0.95 183.3 148.2 1 1.45
NM 7-7 0.95 178.0 147.1 2 1.80
NM 7-8 0.95 183.3 147.1 6 1.76
NM 7-9 0.95 183.3 147.1 1 1.67
NM 7-10 0.95 180.0 144.7 3 1.95
NM 7-11 0.95 186.0 144.7 1 1.66
NM 7-12 0.95 183.3 147.1 5 1.74
KA3 1-2 0.95 170.0 115.4 0 12.35
KA3 2 0.95 170.0 115.4 0 12.92
RK27L2_21 21.1 C 0.75 170 103.6 0 13.10
RK27L2_2121 .8 B 0.75 170 104.6 0 13.49
RK27L2_2121 .5 B 0.75 170 105.6 0 10.72
RK27L2_212 3.2 A 0.7 170 122.8 0 6.61

(continued on next page)

58 L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61

Table A.1 (continued)

Sample ID AI Pm Sm Snf Ф

RK27L2_21 20.4 B 0.7 170 132.0 0 5.82

RK28_23 10.6 A 0.8 170 127.9 0 5.85
RK28_23 10.8 C 0.8 170 140.0 1 6.72
RK28_23 10.9 C 0.8 170 133.6 0 7.42
RK28_23 13.2 A 0.85 170 140.0 0 6.97
RK28_23 10.6 C 0.85 170 140.0 0 5.97
RK30_23 21.0 A 0.8 170 132.2 0 6.84
RK30_23 22.3 B 0.85 170 141.0 0 7.51
RK30_23 22.4 A 0.85 170 143.1 0 6.50
RK30_23 21.1 B 0.85 170 142.0 0 7.47
RK30_23 21.7 B 0.9 170 151.0 0 6.28
Tonalite intrusive NM8-NM8A C2 1 0.8 190.0 122.4 0 3.40
NM9 1-1 0.75 190.0 101.0 1 2.82
NM9 1-2 0.75 190.0 101.9 0 2.70
NM9 1-3 0.75 190.0 101.9 1 3.06
NM9 1-4 0.75 196.7 101.9 0 2.83
NM9 1-5 0.75 196.7 101.9 0 2.99
NM9 1-6 0.75 190.0 106.0 0 2.71
NM9 1-7 0.75 190.0 106.0 0 2.67
NM9 1-8 0.75 190.0 101.9 1 2.87
NM9 1-9 0.75 190.0 95.9 0 2.76
NM9 2-2 0.75 183.3 119.1 0 1.81
NM9 2-3 0.75 183.3 122.1 0 1.67
NM9 2-4 0.75 183.3 123.8 1 1.65
NM9 2-5 0.75 183.3 123.8 0 1.59
NM9 2-6 0.75 183.3 123.8 0 1.57

Table A.2
Values ASI equation (based on the values in Table A.1. for the samples from the Ngatamariki, Rotokawa and Kawerau Geothermal fields (Wyering et al., 2014). ASI = alteration strength
index, UCS (Laboratory) = UCS results from the laboratory testing. UCS (estimated) = The UCS estimates based on Eq. (2).

Sample ID ASI UCS (laboratory) UCS (estimated)

Shallow formations G1 Box 13 1-1 86.57 29.41 20.13

Rhyolitic ignimbrite G1 Box 13 1-2 86.62 33.91 20.18
G1 Box 13 1-3 86.66 28.43 20.22
G1 Box 13 2-1 86.22 13.73 19.78
G1 Box 13 2-2 86.66 21.74 20.22
G1 Box 13 2-3 86.59 25.01 20.15
G1 Box 13 2-4 86.48 28.48 20.04
G1 Box 11 1-1 111.00 49.67 59.58
G1 Box 11 1-2 111.39 50.77 60.51
G1 Box 11 1-3 111.24 35.39 60.17
G1 Box 11 1-4 111.17 57.74 59.99
G1 Box 11 2 112.51 74.69 63.20
G1 Box 17 1 64.58 2.80 5.60
G1 Box 17 2-1 66.99 2.12 6.57
G1 Box 17 2-2 64.53 2.05 5.58
G1 Box 17 3 64.56 2.19 5.59
G1 Box 17 4 64.51 2.21 5.57
G2 Box 15 1 86.48 26.75 20.04
G2 Box 15 2-1 86.49 33.15 20.04
G2 Box 15 2-2 86.52 30.09 20.08
G2 Box 15 3 86.38 30.61 19.94
G2 Box 15 4 86.53 34.56 20.09
G2 1-1 80.17 5.21 14.39
G2 1-2 80.34 7.69 14.53
G2 1-3 80.19 6.66 14.41
G2 2-1 79.78 7.63 14.09
G2 2-2 80.48 9.97 14.64
Rhyolitic lava KAM 11 63.67 23.46 5.26
Siltstone/Sandstone KA30 1-1 99.93 45.38 37.67
KA30 1-2 99.47 29.93 36.92
KA30 2-1 104.34 58.10 45.48
KA30 2-2 104.34 46.43 45.48
KA30 3-1 99.39 34.02 36.79
KA30 3-2 103.56 44.78 44.02
KA30 4-1 103.30 36.60 43.53
KA30 4-2 99.16 28.74 36.41
KA30 4-3 99.07 35.41 36.27
KA30 4-4 99.01 37.27 36.17
L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61 59

Table A.2 (continued)

Sample ID ASI UCS (laboratory) UCS (estimated)

Rhyolitic lava KAW 17 1 86.78 22.11 20.34
Siltstone/Sandstone KAW17 3-2 97.82 32.91 34.31
KAW 17 C2 4 65.83 11.10 6.09
Deep formations NM1 1-2 114.19 80.86 67.44
Rhyolitic ignimbrite NM1 1-1 125.14 114.29 100.58
NM2 2-3 86.05 20.82 19.60
NM2 2-1 90.02 30.95 23.87
NM2 2-2 86.66 29.58 20.22
NM 3 3-2 82.06 22.73 15.94
NM3 3-1 85.82 23.00 19.38
NM3 3-3 80.75 22.87 14.85
NM3 3-7 85.94 24.88 19.50
NM3 3-4 81.01 16.03 15.06
NM 4 100.03 35.29 37.84
NM 4 90.67 22.55 24.64
NM 5 5-2 102.94 62.11 42.88
NM 5 5-1 103.56 39.41 44.02
NM8A C1-1 107.02 51.67 50.81
NM8A C1-2 107.80 62.31 52.44
NM8A C1-3 112.03 65.59 62.05
NM8A C1-5 121.76 100.49 89.27
NM8A C1-6 106.97 53.90 50.71
NM11 1-2 93.68 34.50 28.41
NM11 1-3 93.60 35.32 28.30
NM11 1-4 93.77 35.37 28.54
NM11 1-5 93.80 33.04 28.57
KA37 1-1 101.86 41.22 40.95
KA37 1-2 101.80 39.36 40.84
KA37 1-3 105.05 49.66 46.84
KA37 1-4 105.02 43.75 46.79
KA37 1-5 93.67 28.78 28.40
KA37 1-6 101.77 36.29 40.80
KA37 1-7 101.76 37.41 40.78
KA37 1-8 101.90 34.65 41.03
KA37 2-1 104.84 44.92 46.44
KA37 2-2 105.77 55.47 48.28
Andesite lava/breccia NM 7-1 144.17 130.24 186.60
NM 7-2 146.38 188.08 199.46
NM 7-3 143.82 135.26 184.66
NM 7-4 137.83 106.99 153.34
NM 7-5 116.08 36.72 72.46
NM 7-6 142.41 123.02 176.87
NM 7-7 134.34 113.13 137.10
NM 7-8 111.24 23.92 60.16
NM 7-9 140.74 174.40 167.98
NM 7-10 126.35 127.10 104.89
NM 7-11 138.77 160.83 157.99
NM 7-12 117.15 36.60 75.40
KA3 1-2 109.55 63.19 56.26
KA3 2 109.40 71.37 55.93
RK27L2_21 21.1 C 111.30 69.53 60.29
RK27L2_2121 .8 B 111.90 79.91 61.72
RK27L2_2121 .5 B 113.36 85.99 65.33
RK27L2_212 3.2 A 129.40 105.26 116.43
RK27L2_21 20.4 B 136.05 211.05 144.87
RK28_23 10.6 A 129.32 92.07 116.09
RK28_23 10.8 C 132.35 109.91 128.45
RK28_23 10.9 C 132.69 137.31 129.89
RK28_23 13.2 A 136.29 146.22 146.03
RK28_23 10.6 C 136.93 146.24 149.01
RK30_23 21.0 A 131.92 126.53 126.63
RK30_23 22.3 B 136.81 137.97 148.48
RK30_23 22.4 A 139.07 148.44 159.47
RK30_23 21.1 B 137.65 157.93 152.49
RK30_23 21.7 B 144.67 162.71 189.44
Tonalite intrusive NM8-NM8A C2 1 131.01 123.14 122.88
NM9 1-1 114.73 67.25 68.84
NM9 1-2 120.28 81.72 84.62
NM9 1-3 115.03 65.84 69.63
NM9 1-4 121.72 84.27 89.12
NM9 1-5 121.53 108.24 88.51
NM9 1-6 123.29 83.47 94.24
NM9 1-7 123.34 77.01 94.43
NM9 1-8 115.26 65.56 70.23
NM9 1-9 115.87 76.01 71.89

(continued on next page)

60 L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61

Table A.2 (continued)

Sample ID ASI UCS (laboratory) UCS (estimated)

NM9 2-2 132.74 111.10 130.13

NM9 2-3 135.29 113.09 141.40
NM9 2-4 131.14 87.76 123.42
NM9 2-5 136.76 123.98 148.22
NM9 2-6 136.81 138.11 148.46

References Giggenbach, W.F., 1984. Mass transfer in hydrothermal alteration systems — a conceptual
approach. Geochim. Cosmochim. Acta 48, 2693–2711.
Goff, F., Janik, C.J., 2000. Geothermal systems. In: Sigurdsson, H. (Ed.), Encyclopedia of
Ameen, M., Smart, B.G.D., Somerville, J.M., Hamilton, S., Naji, N.A., 2009. Predicting rock Volcanoes. Academic Press.
mechanical properties of carbonate from wireline logs (a case study: Arab-D reser- Gunsallus, K.L., Kulhawy, F.H., 1984. A comparative evaluation of rock strength measures.
voir, Ghawar field, Saudi Arabia). Mar. Pet. Geol. 4, 430–444. Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 21, 233–248.
ATSM International, 2010. D 7012-10 standard test method for compressive strength and Heap, M.J., Lavallée, Y., Petrakova, L., Baud, P., Reuschlé, T., Varley, N.R., Dingwell, D.B.,
elastic moduli of intact rock core specimens under varying states of stress and tem- 2014a. Microstructural controls on the physical and mechanical properties of
peratures. Annual Book of ATSM Standards. edifice-forming andesites at Volcan de Colima, Mexico. J. Geophys. Res. 1–39.
Baud, P., Zhu, W., Wong, T.F., 2000. Failure mode and weakening effect of water on sand- Heap, M.J., Xu, T., Chen, C.F., 2014b. The influence of porosity and vesicle size on the brittle
stone. J. Geophys. Res. 105, 16371–16389. strength of volcanic rocks and magma. Bull. Volcanol. 76, 1–15.
Baud, P., Vinciguerra, S., David, C., Cavallo, A., Walker, E., Reuschle, T., 2009. Compaction Heap, M.J., Kennedy, B.M., Pernin, N., Jacquernard, L., Baud, P., Farquharson, J.I., Scheu,
and failure in high porosity carbondate: mechanical data and microstructural obser- B., Lavallee, Y., Gilg, H.A., Letham-Brake, M., Mayer, K., Jolly, A.D., Reuschle, T.,
vations. Rock Phys. Nat. Hazards 869–898. Dingwell, D.B., 2015. Mechanical behaviour and failure modes in the Whakaari
Begonha, A., Sequeira Braga, M.A., 2002. Weathering of the Oporta Granite; geotechnical (White Island volcano) hydrothermal system, New Zealand. J. Volcanol. Geotherm.
and physical properties. Catena 49, 57–76. Res. 295, 26–42.
Bibby, H.M., Caldwell, T.G., Davey, F.J., Webb, T.H., 1995. Geophysical evidence on the Henneberger, R.C., Browne, P.R.L., 1988. Hydrothermal alteration and evolution of the
structure of the Taupo Volcanic Zone and its hydrothermal circulation. J. Volcanol. Ohakuri Hydrothermal system, Taupo Volcanic Zone, New Zealand. J. Volcanol.
Geotherm. Res. 68, 29–58. Geotherm. Res. 34, 211–231.
Bieniawski, Z.T., 1967. Mechanisms of brittle fracture of rock part II — experimental Hillier, S., 2000. Accurate quantitative analysis of clay and other minerals in sandstone by
studies. Int. J. Rock Mech. Min. Sci. 4, 407–423. XRD: comparison of a Rietveld and a reference intensity ratio (RIR) method and the
Bieniawski, Z.T., Denkhaus, H.G., Vogler, U.W., 1969. Failure of fractured rock. Int. J. Rock importance of sample preparation. Clay Miner. 35, 291–302.
Mech. Min. Sci. 6, 323–341. Hudyma, N., Burçin Avar, B., Karakouzian, M., 2004. Compressive strength and failure
Binal, A., 2009. Prediction of mechanical properties of non-welded and moderately modes of lithophysae-rich Topopah Spring Tuff specimens and analog models con-
welded ignimbrite using physical properties ultrasonic pulse velocity and point taining cavities. Eng. Geol. 73, 179–190.
load index tests. Q. J. Eng. Geol. Hydrogeol. 42, 107–122. Karakul, H., Ulusay, R., 2013. Empirical correlations for predicting strength properties of
Brantut, N., Heap, M.J., Meredith, P.G., Baud, P., 2013. Time-dependent cracking and brittle rock from p-wave velocity under different degrees of saturation. Rock Mech. Min.
creep in crustal rocks: a review. J. Struct. Geol. 52, 17–43. Sci. 5, 981–999.
Broz, M.E., Cook, R.F., Whitney, D.L., 2006. Microhardness, toughness and modulus of Karfakis, M.G., 1985. Drilling mechanisms at elevated rock temperatures. Int. J. Rock
Mohs scale minerals. Am. Mineral. 91, 135–142. Mech. Min. Sci. Abstr. 22, 407–417.
Ceryan, S., Tudes, S., Ceryan, N., 2008. A new quantitative weathering classification for Kissling, W.M., Weir, G.J., 2005. The spatial distribution of the geothermal fields in the
igneous rocks. Environ. Geol. 6, 1319–1336. Taupo Volcanic Zone, New Zealand. J. Volcanol. Geotherm. Res. 145, 136–150.
Chang, C., Zoback, M.D., Khaksar, A., 2006. Empirical relations between rock strength and Koncagül, E.C., Santi, P.M., 1999. Predicting the unconfined strength of the Breathitt shale
physical properties in sedimentary rock. J. Pet. Sci. Eng. 3-4, 223–237. using slake durability, shore hardness and rock structural properties. Int. J. Rock
Çobanoğlu, I., Çelik, S.B., 2008. Estimation of uniaxial compressive strength from point Mech. Min. Sci. 36, 139–153.
load strength, Schmidt hardness and P-Wave velocity. Bull. Eng. Geol. Environ. 4, Kusznir, N.J., Park, R.G., 1987. The extensional strength of the continental lithosphere; its
491–498. dependence on geothermal gradient, and crustal composition and thickness. Geol.
Coggan JS, Stead D, Howe JH, Faulks 2013. Mineralogical controls on the engineering Soc. Lond., Spec. Publ. 28, 35–52.
behaviour of hydrothermally altered granites under uniaxial compression. Eng. Ladygin, V., Frolova, J., Rychagov, S., 2000. Formation of composition and
Geol. 160:89–102. petrophysical properties of hydrothermally altered rocks in geothermal reser-
Cole, J.W., 1990. Structural control and origin of the volcanism in the Taupo volcanic zone, voir. Proceedings. World Geothermal Congress, Kyushu-Tokoku, Japan. May 28 –
New Zealand. Bull. Volcanol. 52, 445–459. June 10, 2000.
Cumming, W., 2009. Geothermal resource conceptual models using surface exploration Lama, R.D., Vutukuri, V.S., 1978. Handbook on Mechanical Properties of Rocks. vol. 2.
data. Proceedings, Thirty-Fourth Workshop on Geothermal Reservoir Engineering. Editor Trans Tech Publications, Clausthal, Germany.
Stanford University, California. Li, L., Aubertin, M., 2003. A general relationship between porosity and uniaxial strength of
Dinçer, I., Acar, A., Çobanoğlu, I., Uras, Y., 2004. Correlation between Schmidt hardness, engineering material. Can. J. Civ. Eng. 30, 644–658.
uniaxial compressive strength and young's modulus for andesite, basalts and tuffs. Li, Y., Wang, J., Jung, W., Ghassemi, A., 2012. Mechanical properties of intact rock and frac-
Bull. Eng. Geol. Environ. 2, 141–148. tures in welded tuff from Newberry Volcano. Proceedings. Thirty-Seventh Workshop
Dobson, P.F., Kneafsey, T.J., Hulen, J., Simmons, A., 2003. Porosity, permeability, and fluid on Geothermal Reservoir Engineering, Stanford, California.
flow in the Yellowstone geothermal system, Wyoming. J. Volcanol. Geotherm. Res. Luping, T., 1986. A study of the quantitative relationship between strength and pore size
123, 313–324. distribution of porous materials. Cem. Concr. Res. 16, 87–96.
Edlmann, K., Somerville, J.M., Smart, B.G.D., Hamilton, S.A., Crawford, B.R., 1998. Predicting Martin, C.D., 1997. Seventeenth Canadian geotechnical colloquium: the effect of cohesion
rock mechanical properties from wireline porosities. Proceedings SPE/ISRM Eurock, loss and stress path on brittle rock strength. Can. Geotech. J. 34, 698–725.
Trondheim (Norway). Mueller, S., Melnik, O., Spieler, O., Scheu, B., Dingwell, D.B., 2005. Permeability and
Entwisle, D., Hobbs, P., Jones, L., Gunn, D., Raines, M.G., 2005. The relationship degassing of dome lavas undergoing rapid decompression: an experimental determi-
between porosity, uniaxial compressive strength and sonic velocity of intact nation. Bull. Volcanol. 67, 526–538.
Borrowdale volcanic group core samples from Sellafield. Geotech. Geol. Eng. 23, Nara, Y., Meredith, P.G., Yoneda, T., Kaneko, K., 2011. Influence of macro-fractures and
793–809. micro-fractures on permeability and elastic wave velocities in basalt at elevated pres-
Esmaeily, D., Afshooni, S.Z., Mirnejad, H., Rashidnejad-e-Omran, N., 2012. Mass changes sure. Tectonophysics 503, 52–59.
during hydrothermal alteration associated with gold mineralization in the Astaneh Palchik, V., 1999. Influence of porosity and elastic modulus on uniaxial compressive
granitoid rocks, Western Iran. Geochem. Explor. Environ. Anal. 12, 161–175. strength in soft brittle porous sandstone. Rock Mech. Rock. Eng. 32, 303–309.
Fakhimi, A., Gharahbagh, E.A., 2011. Discrete element analysis of the effect of pore size Pola, A., Crosta, G., Fusi, N., Barberini, V., Norini, G., 2012. Influence of alteration on phys-
and pore distribution on the mechanical behaviour of rock. Int. J. Rock Mech. Min. ical properties of volcanic rocks. Tectonophysics 566-567, 67–86.
Sci. 48, 77–85. Pola, A., Crosta, G.B., Fusi, N., Castellanza, R., 2014. General characterization of the
Ferry, J., 1979. Reaction mechanisms, physical conditions, and mass transfer during mechanical behaviour of different volcanic rocks with respect to alteration. Eng.
hydrothermal alteration of mica and feldspar in granitic rocks from south-central Geol. 169, 1–13.
Maine, USA. Contrib. Mineral. Petrol. 139, 125–139. Price, N.J., 1960. The compressive strength of coal measure rocks. Coll. Engl. 37,
Frolova, J., Ladygin, V., Franzson, H., Sigurdsson, O., Stefansson, V., Shustrov, V., 2005. 283–292.
Petrophysical properties of fresh to mildly altered hyaloclasitie tuffs. Proceedings Rawal, C., Ghassemi, A., 2010. Reactive flow in a natural fracture in poro-thermo-elastic
World Geothermal Congress, Antalya (Turkey). rocks. Proceedins: 35th Workshop on Geothermal Reserviour Engineering
Frolova, J.V., Ladygin, V.M., Rychagov, S.N., 2010. Petrophysical alteration of volcanic rocks (Standford, Californa).
in hydrothermal systems of the Kuril-Kamchatka Island Arc. Proceedings, World Reyes, A.G., 1990. Petrology of Philippine geothermal systems and the application of alter-
Geothermal Congress. Bali, Indonesia 25–29 April 2010. ation mineralogy to their assessment. J. Volcanol. Geotherm. Res. 43, 279–309.
L.D. Wyering et al. / Engineering Geology 199 (2015) 48–61 61

Rigopoulos, I., Tsikouras, B., Pomonis, P., Hatzipanagiotou, K., 2010. The influence of alter- for Rock Characterization, Testing and Monitoring: 1974–2006. Turkish National
ation on the engineering properties of dolerites: the examples from the Pindos and Group, pp. 85–92 (1978).
Vourinos ophiolites (northern Greece). Rock Mech. Min. Sci. 47, 69–80. Ulusay, R., Hudson, J.A., 2007b. Standard method for determining the uniaxial compres-
Rowland, J.V., Sibson, R.H., 2004. Structural controls on hydrothermal flow in a segmented sive strength and deformability of rock materials. The Complete ISRM Suggested
rift system, Taupo Volcanic Zone, New Zealand. Geofluids 4, 259–283. Methods for Rock Characterization, Testing and Monitoring: 1974–2006. Turkish
Rowland, J.V., Simmons, S.F., 2012. Hydrologic, magmatic, and tectonic controls on hydro- National Group, pp. 121–132 (1979).
thermal flow, Taupo Volcanic Zone, New Zealand: implications for the formation of Villeneuve, M.C., Diederichs, M.S., Kaiser, P.K., 2012. Effects of grain scale heterogeneity on
epithermal vein deposits. Econ. Geol. 107, 427–457. rock strength and the chipping process. Int. J. Geomech. 12, 632–647.
Sammis, C.G., Ashby, M.F., 1986. The failure of brittle porous solids under compressive Vinciquerra, S., Trovato, C., Meredith, P.G., Benson, P.M., 2005. Relating seismic velocities,
stress states. Acta Metall. 34, 511–526. thermal cracking and permeability in Mt. Etna and Iceland basalts. Int. J. Rock Mech.
Shea, T., Houghton, B.F., Gurioli, L., Cashman, K.V., Hammer, J.E., Hobden, B.J., 2010. Tex- Min. Sci. 7-8, 900–910.
tural studies of vesicles in volcanic rocks: an integrated methodology. J. Volcanol. Vutukuri, V.S., Lama, R.D., Saluja, S.S., 1974. Handbook on Mechanical Properties of Rock.
Geotherm. Res. 190, 271–289. vol. 1. Editor Trans Tech Publications, Clausthal, Germany.
Simmons, S.F., Browne, P.R.L., 2000. Hydrothermal minerals and precious metals in the Walsh, J., 1961. The effects of cracks on the compressibility of rocks. J. Geophys. Res. 70,
Broadlands-Ohaaki Geothermal systems: implications for understanding low- 381–389.
sulfidation epithermal environments. Econ. Geol. 95, 971–999. Wangen, M., Munz, I.A., 2004. Formation of quartz veins by local dissolution and transport
Singh, T.N., Kainthola, A., AV, 2012. Correlation between point load index and uniaxial of silica. Chem. Geol. 209, 179–192.
compressive strength for different rock types. Rock Mech. Rock. Eng. 2, 259–264. Weinberg, R.B., Podladchikov, Y., 1994. Diapiric ascent of magmas through power law
Siratovich, P.A., Heap, M.J., Villenueve, M.C., Cole, J.W., Reuschle, T., 2014. Physical proper- crust and mantle. J. Geophys. Res. 99, 9543–9559.
ty relationship of the Rotokawa Andesite, a significant geothermal reservoir rock in Whitney, D.L., Broz, M., Cook, R.F., 2007. Hardness, toughness and modulus of some com-
the Taupo Volcanic Zone, New Zealand. Geotherm. Energy Sci. 2, 1–31. mon metamorphic minerals. Am. Mineral. 92, 281–288.
Smith, R., Sammonds, P.R., Kilburn, C.R.J., 2009. Fracturing of volcanic systems: experi- Wilson, C.J.N., Houghton, B.F., McWilliam, M.O., Lanphere, M.A., Weaver, S.D., Briggs, R.M.,
mental insights into pre-eruptive conditions. Earth Planet. Sci. Lett. 1-4, 211–219. 1995. Volcanic and structural evolution of Taupo Volcanic Zone, New Zealand: a
Sousa, L.M.O., Suarez del Rio, L.M., Calleja, L., Ruiz de Argandona, R., Rodriguez Rey, A., review. J. Volcanol. Geotherm. Res. 68, 1–28.
2005. Influence of microfractures and porosity on the physic-mechanical properties Wyering, L.D., Villeneuve, M.C., Wallis, I.C., 2012. The effects of hydrothermal alteration on
and weathering of ornamental granites. Eng. Geol. 77, 153–168. the physical and mechanical rock properties of the Andesite Breccia and Tahorakuri
Stringham, B., 1952. Fields of formation of some common hydrothermal alteration min- Formation from the Ngatamariki Geothermal field, New Zealand and empirical
erals. Soc. Econ. Geol. 47, 661–664. relations between rock strength and physical properties. Proceedings, New Zealand
Tamrakar, N., Yokota, S., Sherestha, S., 2007. Relationship among mechanical, physical and Geothermal Workshop, Auckland (New Zealand).
petrological properties of Siwalik sandstones, Central Nepal Sub-Himalayas. Eng. Wyering, L.D., Villeneuve, M.C., Wallis, I.C., Siratovich, P.A., Kennedy, B.M., Gravley, D.M.,
Geol. 3-4, 105–123. Cant, J.L., 2014. Mechanical and physical properties of hydrothermally altered rocks,
Tuğrul, A., Zarif, I.H., 1999. Correlation of mineralogical and textural characteristics with Taupo Volcanic Zone, New Zealand. J. Volcanol. Geotherm. Res. 188, 76–93.
engineering properties of selected granitic rocks from Turkey. Eng. Geol. 51, 303–317. Yagiz, S., 2009. Predicting uniaxial compressive strength, modulus of elasticity and index
Ulusay, R., Tureli, K., Ider, M.H., 1994. Prediction of engineering properties of a selected properties of rocks using the Schmidt hammer. Bull. Eng. Geol. Environ. 68, 55–63.
litharenite sandstone from its petrographic characteristics using correlation and mul- Yıldız, A., Kuşcu, M., Dumlupunar, I., Krrem Aritan, A., Begci, M., 2010. The determination
tivariate statistical techniques. Eng. Geol. 37, 135–157. of the mineralogical alteration index and the investigation of the efficiency of the
Ulusay, R., Hudson, J.A., 2007a. Standard method for determining water content, porosity, hydrothermal alteration on physic-mechanical properties in volcanic rocks from
density, absorption and related properties. The Complete ISRM Suggested Methods Koprulu, Afyonkarahisar, West Turkey. Bull. Eng. Geol. Environ. 69, 51–61.