SOUTHWEST PACIFIC

RIM GOLD-COPPER SYS-

TEMS:
Structure, Alteration and Mineralization

WORKSHOP MANUAL

G J Corbett and TM Leach

Greg Corbett Terry Leach
Corbett Geological Services Terry Leach & Co.
29CarrSt, 54 Ponsonby Rd.
North Sydney, NSW 2060 Ponsonby, Auckland
Australia New Zealand
Ph (61 2) 9959 3060 P h ( 6 4 9) 3 7 6 6 5 3 3
Fax (61 2) 9954 4834 Fax ( 64 9) 360 1010
E-mail [email protected].

SOUTHWEST PACIFIC RIM

GOLD-COPPER SYSTEMS:

Structure, Alteration,

and Mineralization.

Manual for an
Exploration Short Course
presented at Baguio,
Philippines No-
vember 1996

by

G J Corbett &
T M Leach

This Workshop Manual is a minor modification, in the response to reviewers comments, of a presentation for
the SEG/SME at Phoenix, Arizona in March 1996, and is part of an upgrade towards eventual publication,
presumably as part of the Economic Geology Special Publication Series. Additional copies of the pre -
published manuscript are available from Corbett Geological Services above, posted airmail in Australia for
$A75 or overseas $A90.
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

SUMMARY

This workshop classifies differing styles of southwest Pacific rim gold-copper mineralization in
an analysis of hydrothermal ore-forming processes. The magmatic arc geothermal systems in the
Philippines are used as active analogues in the evaluation of the characteristics of intrusive-
related ore systems. Structure and alteration provide information on the direction of fluid flow
within evolving hydrothermal systems, in which we interpret that mixing of magmatic fluids with
meteoric waters provides a mechanism for metal deposition. Major structures localise magmatic
hydrothermal systems in magmatic arc settings and create ore-hosting dilational en-vironments
within subsidiary structures. Breccias occur in most gold-copper deposits and may be categorised
as a guide to understanding the ore-forming environment.

Porphyry copper-gold systems arejocalised, within volcanoplutonic arcs by regional accre-

tionary (arc-parallel) or transfer (arc-normal) structures. Cooling of intrusions emplaced at high
crustal levels results in the initial formation of zoned alteration assemblages, followed by
stockwork veining and the exsolution of volatiles and fluids. Pressure draw-down caused by
cooling of the intrusion and parent melt facilitates the downward percolation of meteoric wa-ters
to porphyry depths, and results in overprinting retrograde alteration. Porphyry copper min-
eralization is interpreted to develop in the apophyses of intrusions by the mixing of meteoric
waters with metal-bearing magmatic fluids derived from larger magma sources at depth. Skarn
deposits exhibit similar prograde and retrograde alteration events and the formation of associ-
ated mineralization, in response to the emplacement of high level intrusions into calcareous
rocks.

High sulphidation gold-copper deposits are derived from magmatic fluids and extend from
porphyry to epithermal regimes. Whereas barren high sulphidation alteration forms as shoul-ders
and caps to porphyry intrusions, more distal mineralized systems are classified as variants of
predominantly structural or lithological control to fluid flow. All systems exhibit character-istic
alteration zonation resulting from progressive cooling and neutralization of hot acid mag-matic
fluid by reaction with host rocks and ground waters. Variations in the style of mineraliza-tion,
metal content and alteration mineralogy, depend on depth of formation and fluid compo-sition. A
two-stage alteration and mineralization model suggests that initial vapour-dominated fluids
develop pre-mineralization zoned alteration, which is overprinted and commonly brecci-ated by
the later mineralized liquid-rich fluids.

Varying styles of low sulphidation gold-copper vein systems predominate in settings of oblique
subduction, where magmatic fluids exolve from intrusive source rocks into environ-ments which
contain meteoric waters of different compositions and temperatures. Quartz sul-phide gold ±
copper systems form proximal to magmatic source rocks by the mixing of mag-matic fluids with
deep circulating cool and dilute meteoric ground waters. Carbonate-base metal gold systems
form at higher levels by reaction of magmatic fluids with low pH bicar-bonate gas condensate
waters. Epithermal quartz gold-silver systems represent hydrothermal systems formed at the
highest crustal levels and display the most distal relationship to the magmatic source. Bonanza
gold grades develop in these systems by the reaction of more dilute magmatic fluids with
oxygenated surficial ground waters. This latter group of deposits is tran-sitional to the classic
adularia-sericite epithermal gold-silver vein systems.

Adularia-sericite epithermal gold-silver deposits form at elevated crustal levels and vary with
increasing depth from: generally barren surficial sinter/hot spring deposits, to stockwork
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

vein/breccias, and fissure veins. Basement metamorphic rocks fracture well and so represent
competent hosts for fissure veins within dilational structural settings. While traditional boiling
models may account for the deposition from meteoric waters of the characteristic gangue min-
erals comprising banded quartz, adularia and quartz pseudomorphing platy carbonate; much of
the gold is interpreted to have been deposited by the mixing of ground waters with magmatic
dominated fluids. Telescoping may overprint the varying styles of low sulphidation gold min-
eralization upon each other or the source porphyry intrusive.

The ore deposit models defined herein are useful in all stages of mineral exploration, from the
recognition of the style of deposit, to the delineation of fluid flow paths as a means of target-
ing high grade ores, or porphyry source rocks. The exploration geologist may be aided by the
use of conceptual exploration models which are interpretative and so vary from the more rig-
orously defined than ore deposit models. Conceptual models should not be applied rigidly but
modified using an understanding of the processes described herein to develop prospect -
specific exploration models.
 Exploration Workshop 'Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

CONTENTS

1. Characteristics of Gold-Copper Hydrothermal Systems

i) Introduction 10
ii) Conceptual exploration models 10
iii) Classification 11
iv) Fluid characteristics 12

2. Geothermal Environment for Pacific Rim Gold-Copper Systems 14

i) Settings of active hydrothermal-geothermal systems 14
ii) Silicic continental and volcanic arc hydrothermal systems 14
iii) Characteristics of Philippine intrusive-related active hydrothermal systems 15
a) Physio-chemical zonations in Philippine geothermal systems 15
b) Waning Stages of active geothermal systems 16
c) Magmatic acid fluid environments 17
d) Analogies to ore-forming systems 17
e) Styles Philippine active hydrothermal systems 18
f) Evolution of active porphyry systems 29
iv) Examples of active intrusion-related hydrothermal systems in the Philippines 20
a) Large disseminated systems in permeable structures or composite volcanic terrains 20
1. Young systems dominated by magmatic vapours 20
2. Circulating hydrothermal systems 21
3. Collapsing hydrothermal systems 21
b) Cordillera-hosted intrusion-related active hydrothermal systems 24
v) Conclusions 25

3. Structure of Magmatic Ore Systems 28

i) Introduction 28
i) Tectonic setting 28
ii) Major structures 30
iii) Framework for dilational structures from active tectonic environments 32
iv) Dilational ore environments 33
v) Fracture systems 37
vi) Shear sense indicators 38
vii) Porphyry and intrusion-related fracture patterns 39
viii) Breccias 42
a) Introduction 42
b) Classification 43
c) Primary non-hydrothermal breccias 44
d) Ore-related hydrothermal breccias 45
1. Magmatic hydrothermal breccias 46
2. Phreatomagmatic hydrothermal breccias 48
3. Phreatic breccias 50
ix) Conclusion 51

4. Controls on Hydrothermal Alteration and Mineralization 51

i) Introduction 51
ii) Temperature and pH controls on alteration mineralogy 51
a) Silica group 52
b) Alunite group 53
c) Kaolin group 53
d) Mite group 54
e) Chlorite group 54
f) Calcsilicate group 54
g) Other mineral groups 55
iv) Alteration zones associated with Ore Systems 55
v) Controls on deposition of gangue phases 56
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

a) Silica 56
b) Carbonates 57
c) Sulphates 57
vi) Controls on metal deposition 57
a) Gold 58
b) Copper 59
c) Lead and zinc 59
d) Silver 59
e) Gold finenes 59

5. Gold-Copper Systems in Porphyry Environments 61

i) Porphyry copper-gold 61
a) Structural setting 61
b) Early models of alteration and mineralization zonation 61
c) Model of polyphasal overprinting events 63
1. Prograde Events 64
i) Heat transfer 64
ii) Stockwork veins 64
2. Retrograde Events 65
iii) Phyllic overprint 65
iv) Mineralization 67
v) Argillic overprint 68
vi) Magmatic high sulphidation overprint 69
ii) Skarn 70
a) Introduction 70
b) Processes of skarn formation 70
1. Prograde isochemical 70
2. Prograde metasomatic 71
3. Retrograde 72
c) Skarn ore deposits 73
iii) Breccia-hosted gold deposit 74
iv) Porphyry and alkaline gold-copper deposits 74

6. High Sulphidation Gold-Copper Systems 76

i) Characteristics 76
a) Classification 76
b) Active analogues 77
c) Alteration assemblages 78
d) Two stage alteration mineralization model 79
e) Mineralization 79
ii) High sulphidation systems formed as shoulders to porphyry intrusions 81
a) Characteristics 81
b) Examples 82
iii) Lithologically controlled high sulphidation gold-copper systems 85
a) Characteristics 85
b) Examples 85
iv) Structurally controlled high sulphidation gold-copper systems 89
a) Characteristics 89
b) Examples 89
v) Composite structurally/lithologically controlled high sulphidation gold-copper 94
a) Characteristics 94
b) Examples 95
vi) Hybrid high-low sulphidation gold systems 98
a) Characteristics 98
b) Examples 98
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

7. Porphyry-Related Low Sulphidation Gold Systems 101

i) Classification 101
a) Introduction 101
b) Sequence of events 102
c) Types of porphyry-related low sulphidation gold systems 102
ii) Quartz Sulphide gold ± copper systems 103
a) Introduction 103
b) Structural setting 104
c) Alteration and mineralization 105
d) Examples 106
iii) Carbonate-base metal gold systems 115
a) Introduction 115
b) Definition 115
c) Distribution 116
d) Geological setting 116
e) Structure 117
f) Alteration and mineralization 117
g) Zonations in vein style and mineralization 118
h) Fluid flow model 119
i) Discussion 120
j) Examples 120
k) Conclusion 133
iv) Epithermal quartz gold-silver systems 134
a) Introduction 134
b) Characteristics 134
c) Structural setting 135
d) Examples 135
1. Associated with intrusion-related mineralization 137
2. Peripheral to intrusion-related mineralization 139
3. Associated with adularia-sericite epithermal gold-silver 140
e) Conclusions 142
v) Sediment hosted gold deposits 143
a) Characteristics 143
b) Example 144

8. Adularia-Sericite Epithermal Gold-Silver Systems 147

i) Classification 147
ii) Examples 148
iii) Tectonic setting 148
iv) Structure 149
v) Fluid Characteristics and hydrothermal alteration 149
vi). Mineralization 150
vii). Types of epithermal gold-silver deposits 151
a) Sinter and hydrothermal (hot spring) breccia deposits 151
1. Characteristics 151
2. Examples 152
b) Stockwork quartz vein gold-silver deposits 153
1. Characteristics 153
2. Examples 153
c) Fissure Veins or Reefs 153
1. Characteristics 154
2. Examples 154
9. Conclusions 159
i) Introduction 159
ii) Gold-copper exploration models in project generation 159
iii) Gold-copper exploration models in reconnaissance prospecting 160
iv) Gold-copper exploration models in project development 160
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

v) If the shoe fits wear it 161

Acknowledgements 162
References cited 162

LIST OF FIGURES

1.1. Pacific rim gold-copper mineralization models

1.2. Southwest Pacific rim gold-copper occurrences
1.3. Size vs grade of some southwest Pacific rim copper-gold occurrances
1.4. Derivation of high and low sulphidation fluids
2.1. Active geothermal systems and hydrothermal ore deposits
2.2. Conceptual model for silicic back-arc rift geothermal systems
2.3. Conceptual model - Volcanic arc hydrothermal systems
2.4. Conceptual Model - Hydrology of shallow levels in geothermal systems
2.5. Geological setting of Philippine geothermal systems
2.6. Tongonan geothermal field - structural setting
2.7. Alto Peak - conceptual hydrological model
2.8. Biliran geothermal field - thermal features
2.9. Biliran geothermal system - conceptual model
2.10. Tongonan geothermal field - conceptual model
2.11. Southern Negros geothermal field - setting
2.12. Southern Negros geothermal field - conceptual model
2.13. Bacon-Manito geothermal field - conceptual model
2.14. Bacon-Manito geothermal field - conceptual model
2.15. Geothermal systems in Volcanic arc-Cordillera settings
2.16. Amacan geothermal system - cross section
2.17. Daklan geothermal field - cross section
2.18. Acupan - geological setting
2.19. Baguio District, Philippines - geology
3.1. Pacific Rim plate margins
3.2. Settings of southwest Pacific rim porphyry copper-gold
3.3. Transfer structures and porphyry systems in PNG
3.4. Structures formed in association with an earthquake in Iran
3.5. Dilational fault systems
3.6. Dilational ore environments
3.7. Riedel's clay model experiment
3.8. Riedel shear model
3.9. Tension gash and domino structures
3.10. Dilational fractures in orthogonal settings
3.11. Fault sense of movement indicators
3.12. Sheeted vein systems - no lateral deformation
3.13. Sheeted vein systems - intrusion under deformation
3.14. Classification of breccia environments
3.15. Magmatic-hydrothermal breccias - subvolcanic breccia pipe
3.16. Magmatic-hydrothermal breccias - structurally controlled
3.17. Magmatic-hydrothermal breccias - injection breccias
3.18. Phreatomagmatic breccias
3.19. Phreatic breccias
4.1. Common alteration mineralogy in hydrothermal systems
4.2. Controls on the solubility of quartz
4.3. Controls on the solubility of calcite
4.4. Controls on the solubility of barite and anhydrite
4.5. Solubility of Au, Cu and Zn relative to temperature and pH
4.6. Controls on the solubility of gold
4.7. Controls on the solubility of zinc, lead and copper
4.8. Gold fineness
5.1. Lowell and Guilbert porphyry copper model
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbet! G J & Leach T M, 8/96 Edn.

5.2. Sillitoe and Gappe porphyry copper model

5.3. Gustafson and Hunt genetic model for the El Salvador porphyry copper
5.4. Pressure-temperature environments at El Salvador
5.5. Porphyry copper model stages I and II
5.6. Porphyry copper model stages HI and IV
5.7. Porphyry copper model stages V and VI
5.8. Porphyry alteration minerals
5.9. Evolution of pluton associated skarns
6.1. High sulphidation systems - styles
6.2. High sulphidation systems - alteration and mineralization
6.3. High sulphidation systems - two stage fluid alteration and mineralization model
6.4. High sulphidation systems - metal zonations
6.5. Horse-Ivaal - plan of alteration
6.6. Horse-Ivaal - cross section of alteration
6.7. Lookout Rocks - plan of alteration
6.8. Lookout Rocks - cross section of alteration
6.9. Vuda, Fiji - structure and alteration
6.10. Vuda, Fiji - conceptual cross section
6.11. Wafi-Bulolo region - structural setting
6.12. Wafi, PNG - plan of alteration
6.13. Wafi, PNG - long section of alteration
6.14. Raffertey's copper-gold conceptual cross section, Wafi
6.15. Nansatsu deposits, Japan
6.16. Miwah - alteration
6.17. Miwah - conceptual model
6.18. Frieda-Nena - setting
6.19. Frieda-Nena - alteration and structure
6.20. Nena - alteration
6.21. Nena - cross section 5200N
6.22. Nena - cross section 4700N
6.23. Nena - alteration long section
6.24. Lepanto/FSE - structural setting
6.25. Lepanto/FSE - geology
6.26. Lepanto/FSE - alteration
6.27. Mt Kasi, Fiji - CSAMT/Structure
6.28. Mt Kasi, Fiji - structure and mineralization open pit.
6.29. Peak Hill - structure
6.30. Peak Hill - paragenetic sequence
6.31. Peak Hill - cross section
6.32. Maragorik - setting
6.33. Maragorik - alteration
6.34. Maragorik cross section
6.35. Bawone-Binebase, Sangihe Is, Indonesia
6.36. Wild Dog - setting
6.37. Wild Dog - geology
6.38. Wild Dog - conceptual cross section
6.39. Masupa Ria - geology
7.1. Low sulphidation gold-copper systems - temporal and spatial zonations
7.2. Low sulphidation gold-copper systems - classification
7.3. Low sulphidation gold-copper systems - alteration
7.4. Ladolam gold deposit - geology
7.5. Ladolam gold deposit - conceptual model
7.6. Kidston - setting
7.7. Kidston - geology
7.8. Kidston - paragenetic sequence
7.9. Kidston - distribution of gangue and ore phases
7.10. Bilimoia - structure
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

7.11. Bilimoia - paragenetic sequence

7.12. Bilimoia - conceptual model
7.13. Arakompa -/paragenetic sequence
7.14. Arakompa -j fluid inclusion data
7.15. Paragenetic/sequence for carbonate-base metal gold systems
7.16. Fluid inclusion data for carbonate-base metal gold systems
7.17. Zonation in carbonate base metal gold systems
7.18. Kelian - geology
7.19. Kelian - fluid flow vectors
7.20. Kelian - carbonate species line 250 E
7.21. Porgera - setting
7.22. Porgera - structure
7.23. Porgera - Waruwari structure
7.24. Porgera - cross section
7.25. Porgera - paragenetic sequence
7.26. Porgera - distribution of carbonate species
7.27. Morobe goldfield- vertical distribution of systems
7.28. Bulolo Graben
7.29. Upper Ridges-Wau diatreme-maar complex
7.30. Kerimenge - composite cross section
7.31. Woodlark Island, PNG- regional structure
7.32. Busai, Woodlark Island - Plan locations of cross sections
7.33. Busai - mineralization cross section
7.34. Busai - alteration cross section
7.35. Busai - paragenetic sequence
7.36. Maniape - structure
7.37. Maniape - paragenetic sequence
7.38. Mt Kare - carbonate-base metal cross section
7.39. Gold Ridge - carbonate-base metal cross section
7.40. Karangahake- cross section
7.41. Porgera Zone VII - paragenetic sequence
7.42. Porgera Zone VII - alteration cross section
7.43. Mt Kare - paragenetic sequence
7.44. Structural setting of the Coromandel Peninsula
7A5. The Thames goldfield, Ohio Creek Porphyry and Lookout Rocks alteration
7.46. Arakompa-Maniape - fluid flow model
7.47. Tolukuma - vein system
7.48. Tolukuma - cross section
7.49. Tolukuma - paragenetic sequence
7.50. Tolukuma - fluid flow model
7.51. Mesel - structure
7.52. Mesel - paragenetic sequence
7.53. Mesel - conceptual fluid flow model
8.1. Models for low sulphidation epithermal vein systems
8.2. Setting of Champagne Pool, Taupo Volcanic Zone, New Zealand
8.3. Golden Cross - structural setting
8.4. Golden Cross - alteration cross section
8.5. Golden Cross - alteration long section
8.6. Waihi New Zealand - structure
8.7. Cracow eastern Australia - structural setting
8.8. Hishikari Japan - structure and alteration

LIST OF TABLES

1. Characteristics of Pacific rim gold-copper mineralization

 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

1 CHARACTERISTICS OF GOLD-COPPER HYDROTHERMAL SYSTEMS i)

Introduction

This is the manual utilised at the short course of the same name presented at the SME/SEG
Meeting in Phoenix in March 1996, with some modifications in response to reviewers com-ments.
The manual is designed so that the figures can be followed during the presentation, in which the
slides of rocks etc further support the concepts delineated herein. I also provides some additional
information not covered in the lectures.

ii) Conceptual Exploration Models

This workshop demonstrates and describes conceptual models as an aid to the exploration and
evaluation of Pacific rim magmatic arc mineral resources. However, we must carefully consider
the nature of these conceptual exploration models before we place any reliance upon them.

As exploration geologists we compare, contrast and classify mineral occurrences in order to build
up empirical patterns from data such as field observations. We develop deposit models as
descriptions of individual deposits, or of more use to the exploration geologist, styles of depos-
its. Exploration models are derived from interpretations, focusing upon those characteristics of a
deposit model which aid in the discovery of ore deposits of a particular style. Progressively
lateral interpretations depart from rigorously reviewed science and so become conceptual explo-
ration models. Such a conceptualisation may give the explorationist a . competitive advantage
(Henley and Berger, 1993) in the increasingly difficult search for ore deposits.

Structure and petrology are tools which the explorationist may utilise in the development of
conceptual exploration models by comparisons of active and extinct hydrothermal systems with
exploration examples. Major structures localise intrusions and minor structures provide ground
preparation. The study of petrology delineates styles of alteration and mineralization, fluid char-
acteristics and mechanisms of ore deposition. The synthesis of structure and petrology may de-
fine fluid flow paths in hydrothermal ore systems. Similarly, models may assist in ranking pro-
jects and aid in the abandonment of lower order targets.

Conceptual exploration models evolve through application to exploration examples and are re-
fined by research, many being abandoned during this process. Although luck plays a part, the
competitive nature of the search for ore bodies encourages explorationists to be the first to de-
velop or utilise a conceptual exploration model. The very innovative nature which makes a con-
ceptual exploration model of use to the explorationist, precludes the lengthy process of rigorous
evaluation of many of the concepts by exhaustive research studies. It is important that models
must not be applied rigidly, but should be modified to become project-specific and great care
must be taken to abandon or modify inappropriate models. It is intended in this workshop to
instruct the participant in the processes involved in the derivation of conceptual exploration
models rather than in the rigid application of existing models.

10
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

iii) Classification

A simple classification is used to distinguish and evaluate differing styles of southwest Pacific
rim gold-copper mineralization (Fig. 1.1, Table 1). Elements of this classification are:

* Crustal level which reflects the proximity to the magmatic source,

* Degree of sulphidation classified as high or low sulphidation as discussed in detail below.

Varying crustal levels of formation provide the primary basis for the distinction of different
styles as:

Porphyry systems are hosted within intrusive rocks at depths of typically greater than 1 km. Cox
and Singer (1988) provide a mean depth of 3.6 km for plutonic copper -molybdenum porphyry
deposits, mainly of the eastern Pacific, and median depths of about 1 km for gold - copper
porphyries typical of the southwest Pacific rim. Sillitoe (1993a) emphasises the vertical extent
(1 km to >2 km) and cylinder shape of the latter deposits. These deposits may contain the
greatest metal contents but at lower grades than deposits formed at shallow levels (Fig.
1.3), and so commonly represent prime exploration targets for bulk low grade mineralization.

The term porphyry is utilised in this manual to describe a high level intrusive rock with a
porphyritic texture, and not necessarily a porphyry copper-gold body in the strict sense.

Mesothermal deposits are described by Lindgren (1922) as "formed ... at intermediate temper-
ature and pressure" and in this classification includes those which developed at temperatures
c
higher than for epithermal deposits, that is >300 C (Hayba et al. 1985). Morrison (1988) also
utilises Lindgen's mesothermal term for veins of the Charters towers district, Eastern Austral-ia,
while Henley and Berger (1993) recognise the difficulties of continuing with the term epi-
thermal for a ranger of deeper deposits such as Kelian, Indonesia. Southwest Pacific rim meso-
thermal deposits are herein described as quartz-sulphide gold ± copper (including Charters
Towers) or carbonate-base metal gold (including Kelian), in order to avoid confusion with the
use of the term mesothermal with slate belt and Mother Lode deposits (Hodgson, 1993), to
which these may be related (Morrison, 1988). The quartz-sulphide gold ± copper and car-bonate-
base metal gold deposits may form resources of considerable size and moderate gold grades
(Fig. 1.3).

Epithermal deposits form at shallow depths and temperatures less than 300°C (Hayba et al.
1985) and encompass a variety of low and high sulphidation deposits. Some display elevated
silver contents and others are characterised by bonanza metal grades exceeding 30 g/t Au (Fig.
1.3), which facilitate extraction by underground mining techniques.

The different styles of southwest Pacific rim gold-copper systems are therefore classified here
as:

* Porphyry-related which includes:

# porphyry copper-gold
# skarn copper-gold
# breccia gold-copper
# porphyry and alkaline gold

* High sulphidation gold-copper. Although commonly described as epithermal in the

geological literature high sulphidation systems extend to the mesothermal and porphyry
regimes, and vary from:

11
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

# barren porphyry shoulders (i.e., on the margins of porphyry systems)

# structurally controlled gold-copper
# lithologically controlled gold-copper
# composite structurally-lithologically controlled gold-copper
# hybrid systems high-low sulphidation gold

* Low sulphidation systems are grouped as:

# porphyry-related deposits which demonstrate the closest relationship to magmatic
source and form a continuum as:
# quartz-sulphide gold ± copper
# carbonate-base metal gold systems
# epithermal quartz gold-silver,
# sediment-hosted gold
# adularia-sericite epithermal gold-silver systems are subdivided with increasing depth
as:
# sinter and hydrothermal breccia gold-silver (hot spring deposits in Sillitoe
1993b)
# stockwork quartz vein gold-silver
# fissure vein gold-silver

Many of these terms are defined below, the characteristics of different deposit types and some
examples are summarised in Table 1.

iv) Fluid Characteristics

The physio-chemical characteristics of the hydrothermal fluids control the:

* type and quantity of metals transported,
* processes which produce mineralization,
* location of the mineralization,
whereas the characteristics of the host rock control the mechanisms of fluid flow (Hedenquist,
1987). A conceptual model for the transportation of fluids from a degassing magma to
porphyry, high sulphidation and low sulphidation systems is illustrated in Figure 1.4.

Country rocks become more competent (brittle) as a result of contact metamorphism during the
initial emplacement of high level porphyry intrusions. Fracturing is initiated at the cooled
margins of the intrusion and extends into the host country rocks. Cooling,of the porphyry in-
trusion and the parent melt is accompanied by the progressive exsolution of dissolved salts,
magmatic volatiles (mainly H2O, SO2, CO2, H2S, HF, and HCl), metals, and their transfer into
the fractured carapace during the evolution of the porphyry systemH(Henley and McNabb,
1978). Dispersion and the local mixing of these magmatic fluids with circulating meteoric-
dominated fluids, results in the zoned alteration and mineralization which characterises porphyry
copper deposits (Henley and McNabb, 1978; e.g., Grasberg and Batu Hijau in Indo-nesia; Ok
Tedi and Panguna in Papua New Guinea). Skams form where mineralizing porphyry intrusions
are emplaced into calcareous host rocks (e.g., Ertsberg, Indonesia; Frieda River Copper, PNG;
Red Dome, eastern Australia).

Volatiles may become overpressured where confined within the intrusion. Tectonic movements
may fracture the carapace and facilitate venting as breccia bodies (e.g., Kidston, eastern Aus-
tralia), and the formation of fracture systems which host later mineralized magmatic fluids.

High sulphidation deposits form if magmatic volatiles and brines are channelled up deep-

12
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

seated fracture/fault zones and rise rapidly with minimal rock reaction or mixing with circulat-
ing meteoric fluids. At temperatures below 400°C the progressive disproportionate of magmat-ic
SO2 into H2S and H2SO4 within the vapour plume produces a hot acid fluid (Rye et al., 1992). As
the temperature decreases, increasing amounts of H2S and H2SO4 are produced (Rye et al., 1992).
These hot acid fluids mix with circulating meteoric waters and react with country rock within
dilational structures or permeable lithologies to form gold-copper deposits (Rye, 1993).
Hedenquist (1987) initially termed these hydrothermal systems "high sulphidation" because
sulphur is in a high oxidation state of +4, due to the dominance of magmatic SO2. However, more
recently (Hedenquist et al., 1994; White and Hedenquist, 1995) thelermH'h'igh sulphida-tion" has
been used to indicate the presence of characteristic alteration and a mineral suite in-cluding
enargite, luzonite and tennantite. The abundance of sulphur cannot be used as a crit eria to
distinguish between high and low sulphidation systems. Sulphur species are commonly abundant
in most, but not all, southwest Pacific high sulphidation systems. Examples of high sulphidation
gold-copper deposits include: Lepanto, Philippines; Nena and Wafi, PNG; Mt Kasi, Fiji; Temora
and Peak Hill, eastern Australia.

In low sulphidation deposits, magmatic fluids which contain dissolved reactive gases are re-
duced by rock reaction and dilution with circulating meteoric waters (Simmons, 1995). The
resultant fluid is dominated by dissolved salts (mainly NaCl) and by H2S as the main sulphur
species. This is interpreted (Giggenbach, 1992) to form at the roots of the low sulphidation
hydrothermal system, where circulating meteoric waters acquire magmatic volatiles and proba-
bly metals. In this case the sulphur is present at an oxidation state of -2 (dominated by H2S) and
was therefore termed by Hedenquist (1987) as "low sulphidation". More recently (White and
Hedenquist, 1995), the term "low sulphidation" has been used to indicate the presence of a
characteristic style of alteration and suite of minerals (such as sphalerite, galena, chalcopyrite)
which form from near-neutral pH fluids. Under these reduced conditions, sulphides are the on-ly
secondary sulphur-bearing minerals with pyrrhotite dominant above 300°C and pyrite at lower
temperatures (Giggenbach, 1987). Examples of low sulphidation gold-copper deposits include:
Lihir and Porgera, PNG; Kelian, Indonesia; Golden Cross and Waihi, New Zealand; Hishikari,
Japan; Kidston, Eastern Australia.

It is interpreted herein that there is an evolution from porphyry to low sulphidation-style flu-ids
through progressive mixing of the magmatic-derived fluids with circulating fluids and wa-ter-
rock reaction. The mixing of low sulphidation mineralized fluids with circulating fluids of
different physico-chemical characteristics produces deposits which are zoned vertically and
horizontally in relation to the source intrusion, from proximal high temperature to cooler distal
settings as: quartz-sulphide gold ± copper, to carbonate-base metal gold, and epithermal quartz
gold-silver. Adularia-sericite epithermal gold-silver systems are form mainly from circulating
boiling meteoric waters and are characterised by the presence of banded quartz, ad-ularia and
quartz pseudomorphing platy carbonate. However, a significant proportion of the gold
mineralization in these systems is interpreted herein to result from the quenching by
groundwaters of circulating fluids which have incorporated the metals from deep magmatic
source rocks.

13
 EPOSIT STYLES EXAMPLES GEOLOGICAL STRUCTURE ALTERATION VEINING PAR- MINERALIZATION
TYPE SETTING AGENESIS
Adularia- Sinter/breccia Osorezan, Cham- fluid upflow zones within dila- brecciated sinter shallow argillic/ polyphasal sinters -> electrum, cinnabar realgar,
sericite Epi- pagne Pool tional settings, advanced argil- veins ->breccias stibnite
thermal Au- lic
Ag
Stockwork/ fis- Hishikari, Cracow, controlled by regional struc- stockwork vein/breccia to deep argil- colloform/crustiform: i) electrum, silver, Ag-
sure veins Golden Cross, Walhi tures varying from fissures grades downwards to local- lic/phyllic and ■ quartz-adularia -bladed cal- sulphosalts/sulphides, chalcopy-
at depth to shallow stock- ly brecciated & banded marginal propylitic cite ii) fine-coarse quartz iii) rite+Au/Ag-tellurides/selenides
works Veins quartz-clay-carbonate iv) clay-
sulphates

Porphyry - Quartz-sulphide Thames, Kainantu porphyry setting controlled banded veins and brecci- phyllic overprint- veining: i) hematite- gold, pyrite, pyrrhotite arsenopy-
Related Low Au+Cu Hamata, Cadia Lake by regional structures, and as controlled by dilational ing propyiit - mgnetite ii) quartz-pyrite - rite chalcopyrite hematite, magnet-
Cowal veins by dilational environ- environment and rock ic/potassic pyrrhotite-As-pyrites iii) chal- ite, Pb-Bi-Cu-Te phases
Sulphidation ments and proximity to the Competency copyrite

Carbonate-base Kellan, Porgera open intrusive phyllic overprint- veining/breccias: i) quartz- gold, pyrite sphalerite, galena,
metal Au pit, Wau, Acupan, ing propylitic adularia/sericite ii) sulphides chalcopyrite, tennantite
Woodlark, Karan- iii) carbonates
Gahake

Epithermal Tofuk'uma, Porgera phyllic/argillic veining/colloform /breccias: gold, pyrite sulphosalts, Au/Ag
quartz Au-Ag Zone 7, Emperor, MI overprinting i) quartz-sulphides ii) tellurides & selenides, Cu-Pb-Zn
Kare propylitic, late quartz-adularia/carb iii) sulphides, hematite
advanced argil- quartz-chlorite-illite
lic

Sediment-hosted Bau, Mesel extensional structures are Disseminated decalcification, vein+breccias: i) pyrite, As-pyrite, arsenopyrite,
gold important dolomitisation quartz-pyrite ii) quartz- stibnite, orpiment, realgar
and silicification As-pyrites
High Sul- Porphyry shoulder Horse Ivaal, regional structures control alteration and mineraliza- zoned potassic, barren to very low grade; covellite-
.ookout Rocks, intrusive emplacement, and tion zonations influenced phyllic, to ad- pyrite +enargite
phidation Structurally Controlled Vuda, Cabang Kid dilational structures host rock by host rock permeability vanced argillic vertically zoned:
Nena, Lepanto, Mt permeability and focus fluid and dilational structures; core silicic, to
Kasi from upflow into outflow zones ore commonly occurs replacement dominated

Lithologically Wafi, Nansatsu as breccia matrix marginal argillic, to i) quartz ii) alunite, covellite, enargite, luzonite,
Controlled Miwah peripheral propylit- barite iii) pyrite tennantite, goldfieldite
ic
Composite Struc- Sahglhe,~Peal< Rid, iv) Cu-sulphides lateral zones: as above outward to
tural and Lithologi- Maragorik tennantite, chalco., base metal
cal sulphides
3
Porphyry anguna, Ok Tedi Gras- regional structure control to Sheeted veins important and early potassic to stockwork: i) quartz- vertical zones: bornite-chalco.-
berg, Batu Hijau intrusive emplacement as fracture mineralization at peripheral propylitic; biotite/K-spar ii) sulphides mag., to chalco.-mag. -pyrite, to
Ertsberg, Ok Tedi splays in accretionary struc- intrusive margins and breccia late phyllic, then iii) sericite-clay-sulphide pyrite-chalco-hem.
Porphyry Cu-Au tures or along transfer struc- matrix infill argillic overprints veining: ) garnet-pyroxene- zoned Cu, to Pb-Zn, to peripheral
tures, subsurface Datholith zoned isothermal, etc. i) oxides-sulphides ii) Au
topography nfluences breccia overprinted by meta- chlorite-carb-quartz
ntrusion somatic, and late
retrograde
Skarn
Breccia Au Kidston, Mt Leyshan as quartz-sulphide Au as quartz-sulphide-Au
Alkaline Porphyry Au Porgera, Lihir potassic, overprinted as quartz-sulphide Au overprinting events: As-pyrite,
by successive phyl- then base metals, then Au-Ag-
lic, argillic and ad- Te phases
vanced argillic

Table 1. Pacific Rim Cu/Au Systems - Summary of Characteristics and Examples

 Exploration Workshop 'Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

2 GEOTHERMAL ENVIRONMENT FOR SOUTHWEST PACIFIC GOLD-COPPER i)

Settings of Active Hydrothermal-Geothermal Systems

Geothermal systems studied over the past decade have provided an increased understanding of
the processes which take place during the formation of hydrothermal ore deposits. Geothermal
systems are encountered in a wide range of geological settings and each one may be analogous
to a distinct style of ore-forming system. These are be classified in terms of their crustal set-ting
and probable heat source (e.g., Henley, 1985a; Fig. 2.1).

Magmatic-sourced geothermal systems occur in association with: oceanic crust along mid -
ocean ridges, ocean island volcanoes formed in relation to hot spots, and back arc basins, or
volcanic arcs along inter-oceanic subduction zones. Exhalative features associated with sea
floor geothermal systems, such as sulphide-rich black smokers, are interpreted to represent
analogies to volcanogenic massive sulphide or Kuroko-style ore deposits (Binns et al., 1993,
1995).

Active hydrothermal systems that have a magmatic heat source may be associated with crustal
rifting within a continental crust, either in back arc rift zones (e.g., Taupo Volcanic Zone, New
Zealand), or in continental rift zones (e.g., East African Rift). As will be shown later in this
section, these types of geothermal systems have a geological setting and fluid chemistry
comparable to the circulating meteoric waters associated with adularia-quartz veining which
host epithermal gold-silver deposits (e.g., Waihi and Golden Cross, New Zealand).

Geothermal systems encountered in volcanic arcs associated with subducting oceanic crust
(e.g., Philippines, Indonesia) are actively forming porphyry-related systems. These systems
form porphyry copper-gold + molybdenum, skarn, high sulphidation copper-gold, and meso-
thermal to epithermal base metal-gold deposits.

Geothermal systems are also encountered in continental environments in the absence of any
obvious magmatic heat source. Rapid uplift results high geothermal gradients which facilitate
the leaching of metals from a thick sedimentary pile by circulating meteoric waters. Fluids
migrate along major fault zones associated with plate collisions (e.g., along the Alpine Fault,
South Island, New Zealand), and deposited gangue minerals and metals in dilational structural
settings as post-metamorphic gold veins (e.g., Macraes Flat, South Island, New Zealand).
Rapid deposition in thick sedimentary basins (e.g., southeast USA) results in the heating of
connate fluids due to overpressuring. These fluids then remobilize metals, forming deposits
such as the Mississippi Valley massive sulphide systems.

ii) Silicic Continental and Volcanic Arc Hydrothermal Systems

There are considerable differences in the geological setting and fluid characteristics between
geothermal systems in silicic continental rift environments (e.g., New Zealand), and in volcan-ic
arc environments (e.g., Philippines; Henley and Ellis, 1983).

In geothermal systems typical of those encountered in silicic rift environments, the heat source
is considered to be a deeply buried (>5-6 km) granite/granodiorite batholith formed from melted
continental crust (Fig. 2.2). Water recharge is derived from meteoric groundwaters and the
intrusion supplies heat, chloride, some gases, and possibly other elements. Boiling occurs at
shallow levels in response to reduced pressure, forming near-surface gas condensate zones. The
upwelling chloride hydrothermal fluid, or chloride reservoir, generally reaches the surface

14
Fig. 2.1

Fig. 2.2
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

as boiling springs, which deposit silica sinters either above the main upflow zone in hydro-
thermal eruption craters (e.g., Champagne Pool, Waiotapu, New Zealand), or in outflow zones
(e.g., Ohaaki Pool, Broadlands, New Zealand). Minor zones of acid sulphate fluids form where
oxidation of H2S occurs above the chloride hydrothermal system.

Active hydrothermal systems associated with volcanic arc terrains display a number of charac-
teristics which are significantly different from those in continental silicic environments (Figs.
2.2, 2.3). In these systems meteoric recharge is typically heated by multiple shallow (<2-3 km)
porphyry intrusions, which contribute significant amounts of magmatic gases, solutes, and
metals to the circulating system. Steeply inclined and elevated volcanic terrains characterised by
permeable pyroclastic sequences, inhibit the upwelling neutral chloride fluid from reaching the
surface above the upflow zone. Instead these fluids flow laterally as far as 5-10 km to low-er
elevations, commonly near sea level. In some cases drilling has not „ encountered circulat-ing
chloride hydrothermal systems until up to 500-1000 m below surface. Condensation of CO2 and
oxidation of H2S in perched aquifers above the circulating system produces extensive reser-voirs
of acid sulphate, bicarbonate and mixed fluids.

Fluid chemistry of the deep circulating hydrothermal fluid is also significantly different be-
tween silicic continental (New Zealand) and oceanic volcanic arc (Philippines) geothermal sys-
tems. Philippine geothermal systems locally contain up to 50 percent magmatic component
(Reyes et al., 1993), whereas New Zealand geothermal systems commonly contain <3-4 per-cent
magmatic component (Hedenquist and Lowenstein, 1994). The fluids in New Zealand systems
are very dilute (<0.3-0.4 equiv. wt percent NaCl), whereas the Philippine systems are almost an
order of magnitude more saline. In addition, the Philippine systems generally exhibit a
significantly higher dissolved gas content. The apparent salinity (actual salinity + dissolved
gases) of Philippine active hydrothermal systems ranges from 2-6 wt percent NaCl, whereas the
apparent salinity of New Zealand systems is generally <1 wt percent NaCl.

A comparison of apparent salinities (NaCl + gas) of active hydrothermal systems and meas-ured
salinities in fluid inclusions in ore deposits illustrates that the fluids in silicic systems in New
Zealand are comparable to those in quartz (-adularia) veining which host epithermal gold-silver
deposits. The volcanic arc active geothermal systems in the Philippines have fluid chemistries
comparable to late-stage mineralized veins associated with porphyry copper, and porphyry-
related carbonate-base metal gold deposits. The following section, investigates these active
porphyry systems in more detail as a guide to the understanding, exploration, and de-velopment
of intrusion-related gold-copper ore systems.

Hi) Characteristics of Active Philippine Intrusive-Related Hydrothermal Systems a)

Physico-Chemical Zonations of Philippine Geothermal Systems

The following four distinct physical and chemical zones have been identified (Mitchell and
Leach, 1991) in active Philippine hydrothermal systems (Fig. 2.3):

1. A Conductive Zone is encountered at depth in many systems, typically within impermeable

intrusives or basement sediments. Heat transfer is predominantly by conduction due to lack of
fracturing. Pressures exceed hydrostatic and approach lithostatic. Where high-level intrusions
have been emplaced into volcanics, impermeable contact nietamorphic assemblages comprise
outer biotite-magnetite and inner clinopyroxene ± amphibole ± biotite. Local skarn assem-
blages are encountered at contacts between intrusives and calcareous sediments (e.g., Palinp i-
non; Figs. 2.11, 2.12).

15
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralization" Corbett G J & Leach T M. 8/96 Edn.

Fig. 2.3

Fig. 2.4
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

2. A Convective Zone forms within the upwelling plume of all Philippine geothermal systems,
within permeable deep-seated fault zones and shatter zones at the contacts of intrusives. The
fluid is commonly slightly less than neutral pH. The alteration exhibits a zonation from:
potassic (predominantly biotite with minor secondary feldspar) alteration at depth, to various
propylitic zones dominated by actinolite, epidote, or chlorite-zeolites at progressively shallower
levels.

3. A Two-Phase Zone is encountered at shallow levels and outflow zones where the total
pressure (hot hydrostatic + gas) exceeds the saturation pressure of water, resulting in exsolution
of water vapour and other gases (mainly CO2, and lesser amounts of H2S and NH3). This gas
exsolution (or passive boiling) can be initiated at depths of 2 km or more in some systems
(Bogie and Lawless, 1987). Within highly permeable zones such as major faults or high-level
dykes, rapid ascent of fluids results in sudden gas exsolution and vigorous boiling. However,
where fluids permeate rocks of uniform, moderate permeability at shallow depths, such as
pyroclastics or jointed lavas with abundant microfractures, gradual exsolution of gases results
in development of a laterally extensive two-phase zone. Condensation of these gases at cooler
levels causes formation of a moderately low pH condensate fluid and associated quartz-illitic
clay-chlorite ± carbonate ± anhydrite alteration.

4. A Phreatic Zone comprises a number of groundwater aquifers perched above the upper level
of the hydrothermal system (Fig. 2.4), and includes the vadose zone (or zone of aeration). Gases
which have evolved from the two-phase zone at depth mix with these perched groundwater
aquifers to produce acid sulphate, bicarbonate, and steam-heated surficial fluids. These fluids
locally reach the surface to form hydrothermal solfataras characterised by amorphous silica,
tridymite and/or cristobalite and where H2S is immediately oxidised to form native sulphur and
minor hydrated sulphates (e.g., halotrichite).

Acid sulphate fluids are formed by oxidation of H2S within surficial aerated groundwater aqui-
fers, mainly above upflow zones, since H2S is rapidly depleted with respect to CO2 in the out-
flow zones. Leaching of host rocks at the site of formation of the acid sulphate fluids, at pH <2,
produces porous silica alteration consisting of amorphous silica, cristobalite, and/or tri-dymite.
Progressive neutralization in response to wall rock reaction and/or mixing with groundwater
forms zones of silica-alunite, silica-kaolinite, and silica-smectite alteration.

Bicarbonate fluids develop by the condensation of CO2 within cool aquifers, both within and
above the chloride reservoir/hydrothermal system. The zonation of Fe —> Mn —> Mg —> Ca
carbonates and kaolinite —> smectite —> chloritic clay alteration reflect progressive neutrali-
zation of the bicarbonate condensate waters. Gypsum is commonly encountered with carbonates
at shallow levels in mixed bicarbonate-acid sulphate waters.

b) Waning Stages of the Active Philippine Systems

As intrusions cool and hydrothermal systems wane, the decrease in temperature and reservoir
pressure results in draw-down of surficial waters deep into the hydrothermal system. Cool, low
pH bicarbonate and acid sulphate waters have been encountered at depths up to 2000 m in some
Philippine geothermal fields (Reyes, 1990b). Down-hole pressure/temperature, geochem-ical,
and stable isotope analyses (Robinson et al., 1987) of these waters have confirmed that they are
derived from perched aquifers in the phreatic zone.

Mixing of cool, descending, low pH bicarbonate-sulphate surficial fluids and hot silica-saturated
deep hydrothermal fluids (Fig. 2.4) results in deposition of carbonates and sulphates (in
response to increasing temperatures) and silica (in response to cooling). Since the

16
 Exploration Workshop "Southwest Padftc rim gold-copper systems: Structure. Alteration, and Mineralization' Corbet! Q J & Leach T M. 8/96 Edn.

hydrothermal systems are invariably located in technically active areas, major fracture/fault
systems may be continually re-opened, permitting descent of surficial fluids to progressively
deeper levels. The overall effect is to seal most permeable features and form impermeable caps
in the upper levels of these hydrothermal systems.

Vertical zonations from gypsum to anhydrite and Fe —> Mn —> Mg —> Ca carbonates
reflect progressive heating and neutralization of descending sulphate and carbonate fluids.
These acid fluids have been encountered at significant depths (e.g., up to 1500 m below
surface at Bacon Manito) where permeable structures have channelled the descent of acid
sulphate fluids, and mineral deposition has isolated these low pH fluids from wall rock
reaction and fluid mixing. The vertical zonations of alunite —> alunite + kaolinite -> py-
rophyllite and/or diaspore in these structures reflects'the progressive heating and neutrali-
zation of the descending acid sulphate fluids.

Cool, dilute, meteoric fluids migrate down major regional structures and provide recharge for
the circulating hydrothermal system. As the system cools and wanes these meteoric fluids en-
croach into hotter regions of the system, producing low-temperature overprinting clay altera-
tion. At cool, shallow levels these fluids contain abundant dissolved oxygen, and are termed
oxygenated groundwater recharge. These fluids are important in the formation of high grade
epithermal gold-silver mineralization.

c) Magmatic Acid Fluid Environments

Fluids within the Philippine geothermal systems are typically slightly less than neutral fluid pH
(5-6 at 250°C) due to significant dissolved gas contents, and are saturated with respect to silica.
However, low pH or acidic fluids are generated under certain conditions, and can play a
significant part in the formation of Pacific rim ore deposits.

As outlined above, acidic fluids may form in aquifers within the phreatic zone, perched above
the main hydrothermal system. These are commonly referred to as steam-heated acid fluids.
Bicarbonate fluids are also formed in the two-phase zone at shallow levels as well as in the
phreatic zone, and are termed condensate acid fluids. Where sulphide-rich alteration zones are
exposed to weathering, oxidation of sulphides can form supergene acid fluids.

Magmatic acid fluids are formed by condensation of SO2 and chlorine gases in magmatic va-
pour plumes which evolve from intrusives at intermediate depths (2-3 km). Where these fluids
directly reach the surface they form magmatic/volcanic fumaroles and solfataras. Geothermal
drilling attempts to avoid intersection of these hot corrosive fluids, although as will be de-
scribed later, exploration of the Vulcan thermal field on Biliran Island encountered magmatic
acid fluids at 1 km depth. In the Bacon Manito geothermal field, topaz and enargite mineraliza-
tion is interpreted (Reyes, 1985) to be indicative of a localised influx of magmatic-rich vola-
tiles into a circulating, meteoric-dominated, near-neutral hydrothermal system. Recent drilling
in the Alto Peak (Reyes ct al., 1993) and Mt Pinatubo (Ruaya et al., 1992) geothermal fields
intersected zones dominated by magmatic volatiles.

d) Analogues to Ore-Forming Systems

The geological setting, fluid chemistry, metal contents, and zonation of alteration and mineral-
ization, indicate that Philippine geothermal systems are analogues to porphyry-related copper-
gold deposits encountered throughout the Pacific rim. Although drilling of active hydrother-
mal systems in the Philippines has not encountered any economic porphyry copper-gold ore.
values of 0.1-0.2 percent copper were identified within potassic alteration at

17
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Palinpinon. In addition, boiling hot water seepages in deep adits at the Acupan gold mine, are
interpreted to represent the last phases of evolution of a porphyry-related carbonate-base met-al
gold system.

Scales deposited from deep chloride reservoir fluids in back-pressure plates in surface pipe-work
from Philippine geothermal systems have graded up to tens of percent of copper, percents of lead-
zinc, thousands of ppm silver, and hundreds of ppm gold (Mitchell and Leach, 1991). Similar
scales in pipework in New Zealand geothermal systems have graded up to percents of gold and
silver (Brown, 1986). Fluids in the New Zealand geothermal systems deposit metals comparable
to epithermal gold-silver deposits, whereas the deeper levels of volcanic arc sys-tems of the
Philippines deposit metals comparable to porphyry copper -gold and porphyry-related carbonate-
base metal gold deposits.

Drilling for geothermal energy in magmatic arc hydrothermal fields in the Philippines has ena-
bled the investigation of porphyry-style systems at depths of greater than 3.5 km below sur-face,
and over areas of up 20-50 km2. Multiple high-level intrusives with associated potassic alteration
zones and local skarn development have been encountered at temperatures of greater than 350°C,
thereby permitting inspection of these potential ore-forming systems during stages of formation.
Detailed petrological work has been carried out on these systems, permitting zo-nations in
alteration and mineralization with fluid chemistry to be compared to pressure-temperature
measurements at depths from which samples were recovered. The formation con-ditions of the
various mineral phases have therefore been empirically determined.

e) Styles of Philippine Active Hydrothermal Systems

The Philippines is a typical volcanic arc setting for porphyry-related hydrothermal systems (Fig.
2.5). Neogene volcanic arcs parallel the Philippine trench to the southeast and the Manila trench
to the northwest, and minor arcs are associated with the Negros and Cotobato trenches in the
southwest.

Active hydrothermal systems in the Philippines are not associated with large stratovolcanoes
(Bogie and Lawless, 1986) such as Mt. Mayon. Volcanic deposits derived from stratovolca-noes
are typically uniform in composition, indicative of a deep (<4-5 km), large, predominant-ly
undifferentiated magma chamber. Circulating meteoric fluids are unlikely to be capable of
penetrating impermeable sediments down to the depths of these intrusives.

Active hydrothermal systems in the Philippines are however encountered in the following two
main geological/tectonic settings (see Mitchell and Leach, 1991):

1. Large hydrothermal systems which are hosted in thick volcanic or volcaniclastic sequences.
These systems occur within composite volcanic terrains parallel to subduction zones (e.g., Alto
Peak, Palinpinon, Bacon Manito), or within favourable structures, such as dilational jogs and
splays related to major faults (e.g. Tongonan, Biliran). These large convective systems form
extensive alteration haloes which are comparable to porphyry-copper/skarn systems.

2. Hydrothermal systems with restricted fluid flow paths are hosted in dilational structural
settings within competent and relatively impermeable basement metasediments and older
intrusions, in Cordillera regions of Eastern Mindanao and Central Luzon.

18
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralization" Corbett G J & Leach T M. 8/96 Edn.

Adapted from Mitchell & Leach 1991 PHILIPPINES

Geothermal Fields
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

0 Evolution of Active Porphyry Systems

Each geothermal system drilled in the Philippines is at a different stage in its evolution, and each
one is a "time slice" into the development of intrusive-related hydrothermal systems. Some
geothermal systems are very young and have probably been drilled shortly after the em-
placement of intrusive heat sources at shallow crustal levels, whereas other systems have been
drilled at late waning stages. The following sequence of events is recorded in the various Phil-
ippine geothermal systems:

1. Contact metamorphism
It is only when active hydrothermal systems evolve sufficiently to exhibit surface thermal man-
ifestations that they become targets for geothermal energy exploration. All active porphyry sys-
tems drilled in the Philippines have already undergone the transfer of heat into the host coun-try
rock, initially by conduction and later by some convection, and formed hornfels in the con-tact
volcanics and sediments (e.g., Tongonan and ABC Man), and zoned skarns in calcareous host
rocks (e.g., Palinpinon, Alto Peak).

2. Magmatic volatile plume

On Biliran Island (Vulcan region) and Alto Peak, plumes of magmatic volatiles are currently
being emplaced into pre-existing zoned skarn and contact hydrothermal alteration assemblag-es.
Down-hole measurements indicate that these magmatic vapours are sourced from a very hot
(>400°C), hypersaline fluid which has exsolved during early crystallisation of the high level
intrusive heat sources. Disproportionation of reactive gases has locally produced hot acid fluids
which have vented directly to the surface as magmatic solfataras (e.g., Vulcan, Biliran Is.). At
depth the acidic fluids have reacted with the host rock to form advanced argillic alteration as-
semblages comparable to those encountered in high sulphidation systems (e.g., Alto Peak,
Palinpinon).

3. Convective hydrothermal alteration

Release of heat and fluids from the high level intrusions establishes deep circulating meteoric
hydrothermal systems into which magmatic fluids are entrained. These circulating systems cre-
ate zoned hydrothermal alteration which grades from an inner potassic zone dominated by bio-
tite to peripheral propylitic alteration. The Tongonan geothermal system is interpreted to be at
this stage of development. The high fluid temperatures (>340-350°C) and salinities (> 15,000
ppm Cl") suggest that a significant input of magmatic brine from the cooling melt has been en-
trained into the convecting hydrothermal system. Although only trace base metal mineraliza-tion
has been deposited from this hot, moderately saline system (Leach and Weigel, 1984), sig-
nificant mineralization has been produced by flashing fluids from depths of >2.5 km to near
ambient conditions within surface pipework (Mitchell and Leach, 1991). It is therefore inferred
that the circulating brine at Tongonan is substantially undersaturated with respect to base and
precious metals. However, the deposition of significant mineralization can be induced under ex-
treme artificial conditions.

4. Meteoric water collapse

Complete cooling of the intrusive heat source produces a pressure draw-down to considerable
depths of cool dilute meteoric waters and shallow, moderately low pH gas condensate (e.g.,
Palinpinon and Bacon Manito). This results in the formation of a zoned phyllic and later argil-lic
overprint on pre-existing contact hydrothermal alteration. The draw-down of these fluids results
in mineral deposition and subsequent progressive sealing of permeable channels at shal-low
levels, and the development of an impermeable cap on the system. Although late stage systems
such as Palinpinon are dilute, the most significant copper mineralization forms in this waning
stage of the hydrothermal system.

19
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

iv) Examples of Active Intrusion-Related Hydrothermal Systems in the Philippines a)

Large disseminated systems in permeable structures or composite volcanic terrains

1. Young Systems Dominated by Magmatic Vapours i)

Alto Peak geothermal System

Alto Peak geothermal system, in the northern Leyte, is hosted in a volcanic arc which extends
along the eastern margin of the Philippine Fault (Fig. 2.6). Drilling within the Alto Peak geo-
thermal system intersected a spatially restricted magmatic vapour plume sourced from a degas-
sing high level intrusion at depth. This vapour plume has been emplaced into a weak, circulat-
ing, moderately saline (7500 ppm Cl.) hydrothermal system, and at depth has resulted in the
formation of a localised advanced argillic overprint on zoned potassic-propylitic alteration. The
following discussion is summarised from Reyes et al., (1993).

The active hydrothermal system at Alto Peak (Fig. 2.7) is hosted in Pliocene to Recent ande-
site-dacite volcanics and subvolcanic quartz diorite dykes, which pass down into a thick se-
quence (>2000 m) of Late Miocene to Pleistocene, locally calcareous, marine sedimentary
breccias, siltstones, mudstones and hyaloclastites (Binahaan Formation). Basement rocks com-
prise Cretaceous harzburgite and pyroxenite.

Composite volcanic centres, domes and collapse calderas are developed within a dilational NW
trending segments (Alto and Cental Faults) of the Philippine Fault System. Additional permea-
bility within the volcanic-sedimentary sequence is also provided by subsidiary EW, NS, and NE
trending faults. Alteration mapping indicates that earlier low temperature clay alteration has
been locally overprinted by vertically zoned
epidote-amphibole-biotite-pyroxene mineralogy, indicative of a later influx of considerably
hotter fluids. Locally, skarns which formed at the contacts with high level quartz-diorite dykes,
display the zonation: garnet-pyroxene —> wollastonite-vesuvianite —> biotite-pyroxene-
amphibole —> quartz-biotite-anhydrite ± epidote, and are considered to be in equi-librium with
current hydrothermal conditions.

Two wells intersected a near vertical magmatic-derived vapour-rich "chimney", 1 km wide and
2-3 km deep, which connects a deep vapour-dominated zone at depth to a shallow zone of steam
heated groundwater. Gas geochemistry, fluid isotope, and fluid inclusion data sug-gests that the
vapour plume contains up to 40-50 percent magmatic component, and is derived from a very hot
(>400°C), saline (>17,000 ppm Cl") fluid. It is interpreted that this fluid has been derived from a
degassing recent intrusion at depth, also the source of the quartz diorite dykes. Alteration at
depth (1700-1800 m below surface) within this magmatic vapour-rich chimney is localised
along fractures and consists of quartz-pyrophyllite-alunite ± diaspore-anhydrite and minor
apatite, zunyite and topaz.

The vapour "chimney" is dominated by CO2 as the main gas phase. The lack of acid C1-SO4 in
the magmatic waters, despite the local occurrence of magmatic-derived advanced argillic al-
teration, has been interpreted to indicate that either the conversion of acidic oxidising magmat-ic
to neutral pH fluids is complete, or it is limited to deeper zones.

ii) Biliran

The Vulcan thermal area is aligned for 3-4 km along a suture zone (Vulcan Fault) within pos-
sible arc-normal structures formed perpendicular to the Philippine trench and cutting the

20
Exploration Workshop 'SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralization* Corbett G J & Leach T M. 8/96 Edn.

Fig. 2.6

Fig. 2.7
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralisation' Corbett G J & Leach T M, 8/96 Edn.

Fig. 2.8

Fig. 2.9
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

island of Biliran, north of Leyte (Fig. 2.8). The presence of abundant native sulphur, superheat-ed
steam, and acid HC1 and SO2 gas condensates, indicate that the Vulcan Fault is directly vent-ing
hot magmatic volatiles. The Biliran system therefore represents an active analogue of high
sulphidation systems. The system is hosted by basement metamorphics which are overlain by 300
m of calcareous sediments and then by a 1.5-2 km thick sequence of andesitic volcanics and
volcaniclastics (Fig. 2.9).

Drilling peripheral to the Vulcan Fault encountered circulating neutral fluids with a significant
fluorine content (an order of magnitude greater than in other fields), indicative o f a substantial
magmatic component. A splay fault from the Vulcan Fault intersected at 1000 m in BN -3 pro-
duced very hot (310-320°C), acidic (pH <2 at depth) fluids.

Down-hole temperatures estimated from alteration mineralogy are lower than actual measured
temperatures in Biliran wells, implying that the hydrothermal system is still heating up. Exso-
lution of hot magmatic volatiles to form high sulphidation systems therefore probably occurs
during the early stages of porphyry evolution. As for Tongonan, there are no volcanics associat-
ed with recent intrusions which might represent the source for heat and magmatic fluid in the
current active hydrothermal system. The active high sulphidation system at Biliran illustrates
strong structural control for the venting of magmatic volatiles from a degassing magma at depth.
Circulating near-neutral hydrothermal convective systems are present outside this struc-ture but
have incorporated a significant component of magmatic-derived fluids.

Drilling at Biliran failed to encounter a significantly permeable or extensive hydrothermal sys-

tem for geothermal exploitation, possibly due to the relatively youthful age of the system.

2. Circulating Hydrothermal

Systems ii) Tongonan

The Tongonan geothermal field is situated in a 12 km long graben formed within a dilational jog
in the Philippine Fault, on the island of Leyte (Fig. 2.6). This jog has been active at least since
the Miocene, controlling emplacement of a number of intrusions ranging from: a large quartz
diorite to granodiorite (10-11 m.y.) pluton between the east and central Philippine Faults, diorite
porphyries (3 m.y.) along the margins of this pluton, and recent felsic porphy-ries and
associated high-level dacite plugs and dykes emplaced along dilational splays cutting the diorite
pluton (Fig. 2.10). The heat source for the currently active system is interpreted to be the recent
stage felsic intrusions which have been emplaced into a thick sequence of Mio-cene to Pliocene
volcanics, and display no associated recent extrusions. 10km to the southeast, the volcanic
centres of Mt. Janagdan and Alto Peak are aligned parallel to the trend of subduc-tion along the
Philippine Trench.

Relatively flat terrain within the graben has enabled the circulating hydrothermal system to reach
close to the surface, with outflows of neutral boiling chloride at lower elevation south to the Bao
Valley. This gentle terrain has facilitated only limited formation of acid sulphate flu-ids above the
hydrothermal system, with minor recharge down the branches o f the Philippine Fault. Alteration
mineralogy in the Sambaloran and Mahiao regions (Fig. 2.11) is generally in equilibrium with
the current hydrothermal conditions, implying that the current system is at least stable, and
possibly waxing. The high salinity of the fluids (up to 16,000 ppm Cl) reflects a high magmatic
component to the system. Argillic to advanced argillic alteration, derived from descending cool
low pH fluids, overprints potassic and inner propylitic assemblages asso-ciated with diorite
intrusions along the branches of the Philippine Fault.

21
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Fig. 2.10

Setting of the Southern Negros Geothermal Field From Mitchell & Leach 1991

Fig. 2.11
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

The Tongonan geothermal field illustrates the strong structural control to multiple intrusive
events and associated overprinting porphyry-style hydrothermal systems. The large dilational
feature which hosts the geothermal system has been reactivated a number of times allowing
emplacement of different generations of intrusions. This strong structural control on multiple
porphyry-style hydrothermal systems at Tongonan is analogous to the tectonic control on in-
trusions associated with the Yandera porphyry copper system, PNG (Titley et al., 1978).

3. Collapsing Hydrothermal Systems

i) Southern Negros Geothermal Area

Two geothermal systems are associated with the Cuernos de Negros volcanic centre, the northern
Palinpinon field and southern Baslay-Dauin field (Fig. 2.11). The volcanic centre comprises twin
summit peaks, dacite domes, and parasitic cones to the south and east, and a py-roclastic plateau
to the north. The most recent pyroclastic'flows (14,000 years BP) are dacitic and were sourced
from a vent 2 km north of the summit craters.

The Palinpinon system is hosted by a 1.5 km thick sequence of Miocene-Recent volcanics which
overlie Eocene-Miocene calcareous sediments and early Eocene volcanics and volcani-clastics
(Fig. 2.12). A large Miocene monzonite pluton was emplaced into the volcanics and sed-iments
in the western portion of the field. Recent porphyry stocks have been emplaced along the eastern
margin of the monzonite, facilitated by movement along a parallel set of north-south trending
dilational structures. These structures have also controlled the venting of the re-cent pyroclastic
flows. The porphyry stocks or deeper equivalent intrusions provide the heat source for the
currently active hydrothermal system.

Two distinct phases of hydrothermal alteration are encountered (Leach and Bogie, 1982): a
relict phase of potassic and advanced argillic alteration in the west, with propylitic alteration to
the east, and a current phase of phyllic alteration in the central regions of the system (Fig. 2.12).
The potassic zone grades from an inner assemblage of clinopyroxene-biotite, outward to zones
of biotite-quartz and actinolite-biotite. In places, zoned hornblende-pyroxene-biotite hornfels
occur at intrusive-volcanic contacts. Elsewhere zoned skarns are intersected within calcareous
sediments adjacent to porphyry intrusions. Alteration grades vertically and eastward from the
potassic zone, through zones of epidote-chlorite, to shallow chlorite-zeolite assem-blages.

Advanced argillic alteration was intersected from surface to 1300-1500 m depth in the western
Sogongon region, and crops out as prominent 2-3 km long ridges aligned north-south along the
main regional dilational structures (Fig. 2.12). The volcanics have undergone alteration to intense
silicification and zoned kaolinite, alunite, and pyrophyllite-diaspore + tourmaline as-semblages.
Trace hypogene covellite mineralization is associated with this advanced argillic alteration.
These zones of intense silicification and advanced argillic alteration are comparable to the
shoulders of high sulphidation alteration formed marginal to porphyry copper deposits (see
Section 6.II), and are interpreted to have formed from comparable magmatic vapour plumes
which are currently venting at Alto Peak and Vulcan, Biliran Is. Similar porphyry-related
silicified ridges extend for 3-4 km along major structures in the Amlan River area 5-6 km north
of the Palinpinon geothermal system, as well as in a belt extending north along the island of
Negros and into Masbate Island (Mitchell and Leach, 1991).

Phyllic alteration at Palinpinon occurs within shallow gas condensate zones, and comprises se-
ricite-quartz-anhydrite + carbonate + chlorite at deep levels, and illitic clay-quartz-

22
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralization' Corbett G J & Leach T M, 8/96 Edn.

Fig. 2.12

Fig. 2.13
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

anhydrite at shallower levels. Phyllic alteration has overprinted the potassic and propylitic
zones, in response to draw-down of gas condensate fluids locally into the deeper portions of the
system. Elsewhere, localised cool acid sulphate-bicarbonate and dilute meteoric recharge fluids
have been encountered at depths up to 2 km, where they have migrated down permeable
structures and along contacts with sills and dykes.

The hydrothermal system at Palinpinon exhibits all the major features encountered in porphyry
copper deposits throughout the southwest Pacific region. These include skarns, homfels, and
potassic alteration zones grading out to propylitic alteration developed during early stages of the
hydrothermal system. Advanced argillic alteration and silicification also formed early in the
evolution of this active porphyry system and are strongly structurally controlled along faults
which facilitated the emplacement of the multiple porphyry intrusions. At present, the Palinpi-
non porphyry system occurs at an early stage of collapse, such that cool gas condensate fluids are
descending in response to pressure draw-down, resulting in phyllic alteration overprint on the
existing zoned propylitic-potassic alteration. The relatively dilute (up to 7,000 ppm Cl) na-ture of
these fluids reflects the minor magmatic input at this stage of evolution of the hydro-thermal
system. Chalcopyrite + bornite are the main copper phases encountered, and copper locally
grades up to 0.1-0.2 percent.

The geological setting of Palinpinon has many similarities to the Panguna porphyry copper de-
posit, Bouganville, PNG (Baldwin et al., 1978), where multiple mineralizing porphyry intru-
sions were emplaced along the margin of a large quartz diorite pluton. Alteration mineralogy in
samples from drilling in the Baslay-Dauin field 6 km to the south, indicates that this system is
youthful and possibly related to a separate series of intrusions at depth associated with the
Cuernos de Negros volcanics.

ii) Bacon—Manito Geothermal Field, Southern Luzon

The Bacon-Manito geothermal field is situated along the Bicol volcanic arc in southern Luzon,
midway between the active large stratovolcanoes of Mayon and Bulusan (Fig. 2.5). The Ba-con-
Manito hydrothermal system is located within a complex composite volcanic terrain of domes,
plugs, and collapsed calderas ranging in composition from dacite to basaltic andesite. Up to 2800
m of volcanics comprising mainly andesites, but ranging from dacite to basalt, overlies a
succession of Miocene to Eocene sediments and volcanics (Fig. 2.13). A number of small stocks
and plugs intersected at depth by drilling in the Cawayan sector vary from pyrox-ene gabbro to
hornblende quartz diorite.

Thermal features occur over a 15-20 km outflow zone and exhibit a vertical zonation common in
volcanic arc geothermal systems. Hydrothermal solfataras occur on the flanks of Mount Pangas
in the region of the main upflow, whereas acid sulphate and mixed sulphate-chloride springs
vent at intermediate elevations. Boiling chloride springs are encountered at around sea level
(Fig. 2.13).

Although the current upflow zone at Bacon-Manito is related to intrusions associated with the
domes of Mount Pangas and Mount Pulog, residual hot spots in the Cawayan region imply that
previous hydrothermal activity was centred on the Cawayan intrusions. Waning of this heat
source has caused draw-down of gas condensate, acid sulphate, and bicarbonate fluids, resulting
in progressive overprinting alteration events (Fig. 2.14). The descent of cool, low pH fluids into
hotter environments has resulted in widespread deposition of carbonates, sulphates, and silica
causing the upper 1000 m to become impermeable. Fingers of acid sulphate and gas condensate
fluids have migrated down structures and caldera margins to depths of over 1500 m. These
fluids have caused extensive phyllic and local argillic overprinting of potassic

23
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralisation" Corbett G J & Leach T M. 8/96 Edn.

Fig. 2.14

Fig. 2.15
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization* Corbett G J & Leach T M, 8/96 Edn.

alteration associated with the emplacement of the Cawayan intrusions.

The multiple high level intrusions, and overprinting potassic-propylitic, phyllic and argillic al-
teration assemblages at Cawayan, are comparable to similar features recorded in the El Salvador
porphyry copper deposit in Chile (Gustafson and Hunt, 1975).

b) Cordillera-Hosted Intrusion-Related Active Hydrothermal Systems

The two main cordillera regions in the Philippines are the Central cordillera in northern Luzon
and the Pacific cordillera in eastern Mindanao. These ranges are composed of uplifted early
Tertiary sedimentary and volcanic sequences, early arc intrusions (typically of Miocene age and
locally containing outcropping porphyry copper deposits), and Mesozoic metamorphic base-ment
complexes. The restricted permeability and recharge at high elevations, together with the
emplacement of shallow intrusions within active volcanic arcs, result in hydrothermal systems
which have characteristics common to both those encountered in flat silicic volcanic terrains and
those in steep volcanic arcs (Fig. 2.15).

Recently-emplaced shallow level plugs and stocks, commonly along contacts of pre-existing
intrusions, provide a heat source for the system. Dacite domes, diatreme breccias and associat-ed
pyroclastic rocks represent extrusive phases to these plugs and stocks. Meteoric recharge
originates at higher elevations than the active hydrothermal system. The restriction of vertical
permeability to structures and dome and diatreme contacts causes the fluid to approach the sur-
face within confined upflow features in which upwelling fluids may boil and mix with de-
scending groundwater, bicarbonate, and acid sulphate fluids. Long outflows via dilational struc-
tures extend towards dilute neutral chloride springs at lower elevations.

These active hydrothermal systems associated with volcanic arcs in cordillera settings are anal-
ogous to porphyry-related quartz-sulphide gold ± copper, carbonate-base metal gold, and epi-
thermal quartz gold-silver deposits.

i) Amacan Geothermal Field

The Amacan geothermal field in north Davao, east Mindanao (Fig. 2.5) is hosted by Mesozoic
metamorphic and Early Tertiary sedimentary rocks which are intruded by multiple intrusions
ranging from early quartz diorite to later microdiorite porphyry related to a Miocene arc (Fig.
2.16). A quartz diorite porphyry and the adjacent metamorphics host the North Davao porphyry
copper deposit.

A number of Pliocene to Recent volcanic plugs and domes are aligned in a north-south trend
parallel to regional structures and to subduction east along the Philippine trench. Lake Leonard is
interpreted to infill a maar/diatreme breccia complex, and is partly infilled and rimmed by
pyroclastic material dated at 1800 years BP (Barnett et al., 1985). The heat source for the ac-tive
hydrothermal system is interpreted to be the dacite plugs and stocks from which the vol-canics
are sourced. Thermal manifestations and circulating hydrothermal fluids are restricted to the
margins of domes and diatremes, with outflows extending 6-12 km northward, controlled by the
north-south regional structures associated with the Philippine Fault. Hydrothermal so l-fataras
are located at the contact of the Ugos dome and around the margins of the Leonard di-atreme.
Neutral chloride springs occur 100 m below the Ugos steam vents, indicating that the circulating
hydrothermal system lies at very shallow depths. Abundant bicarbonate-rich springs in the region
deposit spectacular travertine deposits and reflect the high CO2-content of the hydrothermal
system. Recharge of these bicarbonate fluids play an important role in the formation of
carbonate-base metal gold deposits (see Section

24
 1
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralisation Corbett G J & Leach T M, 8/96 Edn.

Fig. 2.16

Fig. 2.17
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

ii) Daklan Geothermal Field

The Daklan geothermal field, 30 km north of Baguio in the central Cordillera of Luzon (Fig.
2.5), is hosted by Miocene andesite overlying a thick (<2 km) sequence of Early Tertiary vol-
cano-sedimentary breccia which may represent a diatreme breccia (Fig. 2.17). A number of Re-
cent dacite and andesite plugs and domes intrude the volcanic and sedimentary rocks, and are the
site of fumarolic activity. Graben faults channel neutral chloride fluids associated with porphyry
sources interpreted for the dacites. The only permeability encountered during drilling at the
contacts with dykes feeding the Balukbok Dome, provided a relatively saline (2.3 wt percent
NaCl equivalent), and relatively hot (260-270°C) fluid.

iii) Acupan

The Acupan gold mine, 15 km southeast of Baguio (Figs. 2.5, 2.19) is hosted in northeast
trending structures which cut the 1 m.y. Balatoc diatreme and associated dacite plug (Fig. 2.18).
The diatreme and plug (endogenous dome) have been emplaced into a series of Miocene to Pli-
ocene intrusions and andesite volcanics. Hot water seepages are common throughout the upper
levels of the Acupan gold mine, and steam and boiling water discharges occur at deeper levels.
Geothermal drilling encountered minor permeability within the diatreme at around 1000-1200 m
depth.

Fluid inclusion data suggest that the water and steam encountered in the mine are manifesta-tions
of the dying phase of a once more extensive hydrothermal system associated with the gold
mineralization. The styles of alteration and mineralization of the Acupan mine are described in
more detail in Section 7.iii on carbonate-base metal gold deposits.

v) Conclusions

1. Types of Active Porphyry Systems

Active geothermal systems in the Philippines provide valuable insights into the process of
formation of porphyry copper and porphyry-related gold deposits. Drilling to depths of over 3.5
2
km, over more than 20-30 km , enables the development of comprehensive alteration, mineral-
ization, and fluid flow models for these active porphyry systems.

The large porphyry-related geothermal systems at Tongonan, Palinpinon, and Bacon-Manito, are
associated either with permeable regional structures or composite volcanic environments.
Shallow (<2-3 km) multiple intrusions provide the heat source and a significant component o f
magmatic fluids for the circulating hydrothermal system. The nature of host rocks may influ-
ence the style of alteration and mineralization.

If intrusions are emplaced into piles of relatively permeable fractured volcanic rocks, the wide-
spread permeability causes fluids to become dissemination over a significant rock volume,
thereby producing large areas of alteration and mineralization zonation characteristic of most
porphyry copper-gold systems.

If intrusions are emplaced into basement sedimentary, metamorphic or intrusive rocks which
occur within the cordillera regions of the Philippines, hydrothermal fluid flow will be restrict-ed
to permeable structures and intrusive contacts. Extrusions are limited to domes, di-atreme/maar
volcano complexes, and associated pyroclastic material. The confinement of

25
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralisation" Corbett G J & Leach T M, 8/96 Edn.

Fig. 2.18

Fig. 2.19
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

fluids to narrow permeable zones permits fluids to reach near surficial levels where they inter-act
with shallow groundwater and abundant bicarbonate and sulphate waters. This type of ac-tive
hydrothermal system is analogous to the intrusion-related gold deposits encountered throughout
the southwest Pacific rim (see Section 7).

Contrasts in permeability localise different styles of high grade deposits at Hishikari, Japan and
Grasberg, Indonesia. In each case competent low permeability rocks have fractured well to host
the mineralization and are overlain by permeable volcanic rocks which facilitate the downward
movement of hydrothermal fluids which promote mineral deposition. Grasberg is interpreted to
have been emplaced into a diatreme breccia by MacDonald and Arnold, (1994) and the Hishi-
kari fissure veins are capped by altered volcanic rocks (Section 8.b.iv).

2. Evolution of Active Porphyry Systems

The geothermal systems in the Philippines have been investigated at various time slices in the
evolution of intrusive-related hydrothermal systems. In each case, the initial emplacement of
high level intrusions has resulted in conductive transfer of heat during the formation of contact
metamorphic hornfels or skarns. Subsequent cooling of the intrusion has been accompanied by
exsolution of fluids and formation of magmatic vapour plumes. Fluid flow may be controlled by
the same dilational structures which may have facilitated the emplacement of the intrusion.
These magmatic vapours vented to the surface as solfataras at Vulcan, and resulted in an ad-
vanced argillic overprint to the earlier alteration at Alto Peak.

Convective hydrothermal systems develop when the heat provided by an intrusion at depth, and
possibly also venting of magmatic fluids, results in the formation of circulating meteoric -
dominated hydrothermal systems at higher crustal levels. Early convective systems are poorly
developed and have a high magmatic component (e.g., high fluorine at Biliran), and later ones
are well developed and saline (e.g., Tongonan). The convective transfer of heat from the source
intrusion results in the formation of zoned potassic-propylitic alteration.

Cooling of the source intrusion and associated pressure draw-down in the hydrothermal sys-tem
causes the descent of cool relatively low pH gas condensate and surfical acid sulphate-
bicarbonate waters (Palinpinon and Cawayan respectively) to depths up to 1.5 km below the
surface, and overprinting phyllic and argillic alteration. The deposition of sulphates, carbonates
and quartz from the descending waters causes sealing of permeable zones and accelerates the
collapse of the hydrothermal system.

3. Porphyry Copper and Epithermal Gold Systems

The close proximity of many Philippine gold deposits to porphyry copper deposits has led
many authors to speculate that the gold systems are formed at shallow levels within the same
porphyry environment (e.g., Sillitoe and Bonham, 1984). Epithermal gold systems form above
and peripheral to porphyry source rocks and telescoping locally juxtaposes these. However in
many cases, Philippine porphyry copper-gold deposits are of Miocene age and the epithermal
gold deposits of generally Pliocene or younger ages. These deposits are interpreted to have
formed in association with arcs of different ages and are separated by an episode of uplift and
erosion (Mitchell and Leach, 1991).

Similar factors may localise but systems in the same geological setting. Porphyry copper-gold
and epithermal gold deposits occur at the same stratigraphic level, i.e., at or slightly above the
interface between andesrtic volcanics and underlying basement sediments. Reactivation of the
same structural environment may also contribute to the overprinting.

26
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

3 STRUCTURE OF MAGMATIC ORE SYSTEMS

i) Introduction

Major regional crustal structures act as conduits for rising porphyry intrusions which are the heat
sources for hydrothermal systems. In southwest Pacific settings of oblique_subduction, flu-ids
exsolve from intrusive (porphyry) source rocks and minerals deposit in peripheral settings as
veins, fractures, and breccias. Movement on the major controlling structures commonly cre-ates
ore-hosting dilational environments in high-angle subsidiary structures. If drilling is planned on
grids normal to the more obvious major structures, then in many instances, drill holes intersect
the actual mineralized features at unsatisfactorily low angles to the core axis (Fig. 3.4).

Very large quantities of fluid may be required to transport economic quantities of gold. At
solubilities of 10 ppb (Fig. 4.6), 10 *~7 litres of fluid (or about equivalent to an Olympic
swimming pool) would be required to transport about 3 oz. of gold (Brown, 1986). Optimal
mineral deposition is interpreted to be best promoted at the sites of mixing of magmatic and
ground waters (Section 4.v.a). The formation of economic gold deposits requires the channel-
ling of considerable quantities of fluid through otherwise impermeable host rocks. Brittle frac-
turing within competent host rocks in the upper portion of the crust provides permeability for
the formation of magmatic arc gold-copper mineralization. Host rocks fracture within dilation-al
structural environments and act as both fissure conduits for mineralized fluids, and as set-tings
for mineral deposition. Fracture-permeability is an important mechanism of fluid flow in
pervasive alteration.

The theme - major structures localise ore systems in which ore is hosted within subsidiary
high angle structures - is common in many southwest Pacific rim gold-copper systems.

ii) Tectonic setting

Most Pacific rim gold-copper mineralization is associated with Tertiary subduction-related

volcanoplutonism, commonly within island arcs which develop at convergent plate boundaries
(Sillitoe, 1992). Varying styles of convergence are distinguished by Sillitoe^Q^2) as:
1. Orthogonal convergence,
2. Oblique convergence,
3. Arcs associated with rifting
4. Back-arc extension.

Some terrains exhibit changes in subduction style through time, (e.g. Luzon, Philippines; East
New Britain, PNG). We suggest that changes in the style of subduction mayjnitiate_pi)jphyry_
intrusion.and particular .tectonic settings promote the development of certain deposit types.

1. Orthogonal convergence is characterised by orthogonal plate collision and the formation of

an island arc in the overlying plate (e.g., northern Chile; Banda Arc, Indonesia; East New
Britain, PNG; Fig. 3.1). Some arcs change character to exhibit orthogonal convergence during a
limited time (e.g, Northern Luzon, Philippines in the Miocene; Sillitoe, 1992). In these ter-
rains, fluids are constrained within the intrusion and so porphyry copper-gold deposits devel-op
with only minor peripheral vein systems.

27
 Exploration Workshop 'SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralisation" Corbet! G J & Leach T M, 8/96 Edn

See also figure 1-2 for detail of the SW Pacific Rim.

 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

2. In settings of oblique, convergence, the plates slide past each other and major transcurrent
fault systems accommodate much of the displacement (e.g., Philippine Fault, Philippines, Fig.
2.4; Sumatran Fault System, Indonesia, Fig. 1.2; Alpine Fault, New Zealand, Fig 7.44; Queen
Charlotte Fault, Canada). There is increasing evidence that regional strike, slip structures
localise porphyry sy§tems_at splays or jogs (e.g., Philippine geothermal systems, section 2;
section iv.b.l herein). Settings of oblique rather than orthogonal convergence, are also liable to
develop dilational structural environments suitable for the formation of mesothermal porphyry-
related mineralization formed peripheral to the porphyry environment. Rising fluids readily
exsolve from porphyry environment to form both low and high sulphidation gold mineralization
(e.g., Lombok Tandai and other deposits along the Sumatran Fault, Indonesia; Thames
goldfield, New Zealand, Fig 7.45; Nena, PNG, Fig. 6.18; Lepanto, Philippines, Fig.
6.24; Cracow, eastern Australia, Fig. 8.7). The dilational character of individual Ordo vician
porphyry-related deposits in New South Wales, Australia suggests that mineralization formed
when the arcs dispayed an oblique character (Corbett unpubl. data).

Extensional environments in subduction-related arc settings are described by Sillitoe (1989) as

important but variable settings for gold mineralization. There are considerable differences in the
manner in which workers categorize southwest Pacific rim extensional settings of mineral-
ization and so further study is required.

3. Intra-arc rifts represent sites of crustal thinning and volcanism. Many workers (Sillitoe,
1992; Mitchell and Garson, 1981) describe the Green Tuff Belt of Japan, which hosts Kuroko
deposits as massive sulphide deposits as of this setting. Elsewhere, localised intra-arc rifts may
host mesothermal carbonate-base metal gold mineralization in association with dacitic
intrusions and phreatomagmatic eruptions. Rotation on transfer structures has formed the
Bulolo Graben, PNG, as a setting of crustal thinning and high level volcanoplutonism which
hosts the Morobe goldfield (Fig. 7.28, Corbett, 1994). Some 3.7 M oz of (mainly alluvial) gold
has been produced at Morobe, and published hard rock reserves of 5.5 M oz remain, localised
by graben-bounding and intra-graben faults (Section 7.iii.j). A jog in the Hauraki Fault, which
separates the Coromandel Peninsula from the Hauraki Graben, hosts the Ohio Creek porphyry
and Thames epithermal goldfield (Figs. 7.44, 7.45).

4. Back-arc extensional environments are described by Sillitoe (1992) as an importantsettings

for adulaffa^sericrfe epithermal gold-silver deposits,. In New Zealand, the back-arc
environment of the Taupo Volcanic Zone (Fig. 8.2) provides a modern analogy for the setting
for Miocene gold mineralization (Fig. 7.44) of Coromandel Peninsula (Henley, 1985b; Henley
and Hoffman, 1987). The terms ensialic marginal basin (Cole, 1984) and intra-arc rift (Sillitoe,
1992) are also used to describe the Taupo Volcanic Zone. In back -arc settings, as typical for
adularia-sericite epithermal gold-silver deposits, magmatic,heat sources are inferred to occur at
considerable depth and mineralized vein systems are hosted by competent host rocks: In Japan,
the apparent Pliocene to younger ages of many adularia -sericite epithermal vein systems,
association with major structures and basement host rocks, are all typical of a back-arc
environment (e.g. Hishikari, Rubeshibe belt in Hokkaido). Many of the major structures are
reactivated rifts associated with the Green Tuff Belt.

Intrusion-related deposits are also apparent in back-arc settings. The caldera-hosted Emperor
Gold mine, Fiji (Eaton and Setterfield, 1889), classified herein as of the epithermal quartz
gold-silver style, and the Marian and Didipio porphyry copper-gold prospects, Philippines
occur in back arc settings (Sillitoe, 1992).

28
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Hi) Major structures

The spatial relationships between major structures and mineralization through geological time
are well documented (O'Driscoll, 1986). In magmatic arcs regional structures may localise in-
trusions and also create dilational structural environmetns which host ore. Major structures may
demonstrate protracted histories of activity as: pre-mineralization controls on basin sedimenta-
tion in host rocks, pre-mineralization intrusion or breccias, through syn-mineralization locali-
zation of ore systems, and post-mineralization deformation of ore deposits. Pre-existing frac-
tures systems are common ore environments.

Major structures in subduction-related Pacific rim settings (Figs. 3.1, 3.2) may be classified as:

1. Accretionary
2. Transfer structures
3. Conjugate transfer structures
4. Transform faults

1. Accretionary (arc-parallel) structures form parallel to the subducting plate margin and in part
define the structural grain of the accretionary prism. In settings of oblique subduction, arc-
parallel structures may form as steeply dipping transcurrent faults which display predominant-ly
strike slip components of displacement (e.g., Philippine Fault, Philippines; Sumatran Fault
System, Indonesia). Differing styles of deposits may be exposed at varying levels of erosion. At
the deepest levels of erosion, the Domeyko or Western Fault in Northern Chile localises many
porphyry systems (Boric et al., 1990; Davidson and Mpodozis, 1991), and the giant La
Escondida porphyry was identified by prospecting along that structure (Lowell, 1991b). At
mesothermal levels, differing styles of gold deposits are localised by the Gilmore Suture, east-
ern Australia, and a structural corridor in Kalimantan (van Leeuwen et al., 1992; Fig. 1.2). Epi-
thermal gold mineralization forms in structural environments created by the Sumatran Fault
System, and a structural corridor in Southern Hokkaido termed the "Tokuryu Structural Zone
(TSZ)" (Corbett, unpubl. data, 1987). The Tongonan geothermal field is related to active
porphyry systems localised by the Philippine Fault (Fig. 2.5).

While many accretionary structures may be identifiable as one or more individual crustal dis-
continuities, others occur as sutures which separate segments of accreted terrains within mag-
matic arcs. The Kalimantan Suture represents a zone of varying styles of porphyry-related gold
mineralization recognised by van Leeuwen et al., (1990). This structural zone forms the ter-rain
boundary between the Kalimantan Magmatic Arc of Mitchell and Carlile (1994) to the west and
mainly sedimentary cover to the east (van Leeuwen et al., 1990). Many high leve l porphyry
intrusions and gold occurrences occur along this corridor (Fig. 1.2; Pieters and Su-priatna,
1990). Similarly, the Gilmore Suture (Stuart-Smith, 1991) separates rocks of differ-ing ages. The
Carlin Trend (Bonham, 1988) is another well known linear belt of deposits.

Cross structures and dilatant zones localise porphyry-systems within accretionary structures,
which may extend for hundreds of kilometres. Dilatant zones which develop as flexures or jogs
(see Section 3.iii) may display a configuration of flower structures (Lowell, 1987) in pro-file.
Where eroded to porphyry levels these are evident as splays which localise porphyry sys-tems
(e.g., the porphyry systems at FSE, Philippines, Fig. 6.22, Baker, 1992; Frieda River, PNG, Fig.
6.18; Corbett, 1994; Chuquicamata, Chile, Boric et al., 1990), and occur as pull-apart basins or
jogs at much higher levels into which intrusions are emplaced at depth (e.g., Tongonan
geothermal field, Philippines, Fig. 2.5). Gold mineralization occurs in association

29
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralization" Corbett G J & Leach T M. 8/96 Edn.

Fig. 3.2

Fig. 3.3
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

with pull-apart basins (e.g., Sumatran Fault, Indonesia at Mangani, Kavalieris et al., 1987; and
Lampumg, Corbett, unpubl. report, 1993).

2. Transfer (arc-normal) structures accommodate the along-strike or dip variations in linear

detachment faults (Etheridge et al., 1988), and have been expanded using the classification of
Lowell (1987) to also include collision settings. Where at high angles to the accretionary
prism, these structures may separate segments of a subducting slab and so accommodate
variations in the dip or rate of subduction (Fig. 3.2). They may act as deep crustal
discontinuities to facilitate the emplacement of melts derived from considerable depths into
settings at higher crustal levels (e.g., Porgera and Mt Kare are localised by the Porgera
Transfer Structure, PNG, Figs. 3.3, 7.21; Corbett, 1994), or focus overprinting intrusions (e.g.,
Grasberg, Indonesia; Wafi, PNG, Fig. 6.11). Porphyry systems are commonly localised at the
intersections of transfer and accretionary structures. As deep basement features structures,
transfer structures commonly cannot easily be seen to displace folded and thrusted cover rocks
within accretionary prisms, but may be identified as fractures which penetrate the cover rocks
most readily discernible on quality Landsat imagery or aeromagnetic data (Corbett, 1994).

3. Conjugate transfer structures intersect accretionary prisms at moderate angles and may
localise porphyry intrusions at the intersections with accretionary structures. Although formed
in settings of orthogonal compression, (e.g., Banda Arc, Indonesia; and Northern Chile), these
structures may be reactivated as strike slip structures during localised extension at a relaxation
of compression. Conjugate transfer structures commonly host mesothermal vein mineralization
formed peripheral to porphyry intrusives. The strike slip rotation displayed by these structures
may facilitate the formation of higher grade portions of vein systems within localised dilational
zones (Fig. 3.9).

4. Transform faults are defined as strike slip faults which terminate against major tectonic
features such as plate boundaries and commonly separate segments of oceanic spreading
centres (Wilson, 1965; Biddle and Christie-Blick, 1985). Fractures which separate and offset
segments of oceanic ridges are well documented (Fig. 3.1). The San Andreas Fault System,
USA, (Fig. 3.1); Queen Charlotte Fault, Canada, and the Sumatran Fault System, Indonesia
(Fig. 1.2), are all major faults which terminate against spreading centres. The youthful San
Andreas transform Fault System is poorly eroded and so localises gold mineralization formed
at surficial crustal levels at McLaughlin (Tosdal et al., 1993). The more deeply eroded Queen
Charlotte Fault localises the Cinola gold deposit (Corbett, unpubl. data, 1993), and the Sumatra
Fault System hosts many epithermal gold-silver deposits (Fig. 1.2). Rotation on transform
faults has created dilational structural environments which host high sulphidation alteration in
East New Britain, PNG (Fig. 6.32).

iv) Framework for dilational structures from active tectonic environments

Seismic activity, such as earthquakes, results from the interaction of the moving plates about the
Pacific rim (Fig. 3.1). In settings of oblique subduction much of the displacement, as the plates
slide past each other, may taken up by transcurrent faults. The San Andreas Fault in California
(Fig. 3.1) displays some 260 km of strike slip displacement (Crowell, 1962), as a transform
fault which separates two segments of oceanic spreading centres (Wilson, 1965). Earthquake
epicentre locations indicate that the same major structures, such as the San Andre-as Fault, have
been reactivated during successive earthquakes (Sibson, 1989).

Modern analogues of plate motions, which impart distinguishable and consistent strike-slip
fault rotations at the frequency of earthquake activity, contribute towards a model of ore

30
Exploration Workshop -SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralization" Corbett G J & Leach T M. 8/96 Edn.
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

formation. Strike-slip structures are rarely single, straight, linear features, but contain irregularities.
During strike-slip deformation irregularities such as bends, or zones where movement changes from one
structure to another, may represent sites of compression or dilation (opening). An earthquake at Dasht-e
Bayaz, Iran in 1968 provides an excellent example of dilation in such a setting (Fig. 3.4, Tchalenko and
Ambraseys, 1970). Here, the displacement of cultural features delineates the direction of movement on a
strike-slip fault. Subsidiary structures which link the main faults form a jog be-tween the two major
structures. These are analogous to dilatant fissure vein ore systems. Differing styles of dilatant features
are described in the literature by a variety of terms (Figs. 3.5, 3.6; Segall and Pollard, 1980; Sibson,
1989). It is important to note the configurations in which subsidiary struc-tures may be dilational or
compressional, depending on the direction of rotation of the controlling ma-jor structure. Dilational
settings within fault systems are important mineralizing environments (Sibson, 1986, 1987, 1992).

Following earthquake rupture, dilational fault systems may become the sites of enhanced flow of flu-
ids derived from considerable depths by a suction pump-like mechanism (Sibson 1987, 1991, 1992). In
this model, faults in the active geotectonic Pacific rim settings may display regular displacements with
consistent directions. Each time an earthquake movement promotes development of a dilational zone,
fluids are squeezed out of the enclosing rocks into the lower pressure regime within the dila-tional
structure (Fig. 3.5). Hot fluids will be forced up the dilational zone and may deposit minerals by:
boiling in response to instant depressurisation (flashing), cooling, or mixing with meteoric waters.
Typical banded epithermal veins and dilational breccias develop by the deposition of multiple bands of
minerals such as quartz and adularia. While boiling of meteoric waters (Henley, 1985a) is favoured as a
mechanism for the deposition of quartz, adularia and platy carbonate (commonly subsequently
pseuodmorphed by quartz) in epithermal vein systems, gold derived from magmatic sources may be
deposited by mixing (see Section 4.vi.a). The important feature of this model is that dilatant fracture
systems provide mediums for hydrothermal fluid transport and that repeated fault activation is the
mechanism which may produce higher grade ores in many epithermal vein systems characterised by
banded veins. Thus, it is possible to transport large quantities of fluids required for the formation gold
deposits.

v) Dilational ore environments

In settings of oblique collision much porphyry-related gold-copper mineralization displays associa-

tions with inferred dilational features within regional strike slip structures, (e.g., deposits along the
Sumatran Fault and Philippine Fault). Similarly, there is an increasing body of evidence that in set-
tings of orthogonal collision and subduction, porphyry intrusion occurs during periods of extension,
developed as a result of relaxation of subduction.

a) Aspects

Important aspects of dilational ore systems are:

1. Angular relationship between controlling and dilatant structures

2. Host rock competence.

1. Angular relationships of dilatant tension gash fractures to the controlling fractures should be careful-
ly resolved in planning the orientation of exploration drilling or trenching (Figs. 3.5, 3.6). The re-gional
strike-slip structures which localise an ore system may be mappable over considerable dis-tances and
contain some vein mineralization. However, the dilatant tension

31
Exploration Workshop 'SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralisation' Corbett G J & Leach T M, 8/96 Edn.
Fig. 3.5

Fig. 3.6
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

gash veins are initiated at angles of 45° to the controlling structures and are progressively ro-
tated towards 90° during the strike-slip deformation on the regional structure with which min-
eralization is associated (Fig. 3.9). Tension gash veins at high angles (approaching 90°) to the
controlling structures display the greatest dilation and so commonly exhibit greater thicknesses
and higher gold grades. Thus, drill testing normal to the orientation of the controlling struc-ture
could be at very low angles to, or essentially parallel to the mineralized gash veins. Erratic
results will include anomalously high assays from the few drill holes which might bore down
individual veins, while many more may be bored between veins and so will be barren. These
problems could be further compounded when vertical drill patterns are used to test vein sys-tems
which commonly display subvertical orientations.

2. Competent host rocks which fracture well are required for:

* the formation of fissure-style vein systems,

* fracture controlled permeability in hydrothermal systems.

Fissure vein-style gold-silver deposits form within large-scale tension fractures (discussed
below) hosted by competent rocks. In Japan, although epithermal gold deposits occur in the
Miocene Green Tuff terrain, basement shales host the fissure vein gold mineralization at
Hishikari (6.8 M oz Au) and Konami (2.35 M oz Au), while competent intrusive domes em-
placed into volcaniclastic sequences host vein systems at the Sado (2.5 M o z Au) and Chitose
(0.9 M oz Au) deposits. At Hishikari, the pronounced contrast between the competent Shimanto
Group shale host rocks to the ore and the overlying clay altered and hence incompetent vol-canic
breccias has constrained mineralization within a 100 m vertical interval at the top of the
basement Shimanto Group shales, to produce gold ores of a 80 g/t gold head grade and local
bonanza gold grades (Section 8.vii.c.2). In the Coromandel Peninsular of New Zealand, the
major fissure vein systems at Martha Hill (5 M oz Au), Golden Cross (1 M oz Au), and Ka-
rangahake (4 M oz Au), are hosted within massive andesite lavas while rhyolitic pyroclastics are
poorly mineralized. The fissure vein at Karangahake passes to a subeconomic stockwork vein
system at the transition from the andesite to the overlying rhyolitic pyroclastic rocks.

Fracture permeability provided by small-scale fracture networks such as crackle breccias (Sec-
tion 3.viii.c.l) allows fluid flow for the development of hydrothermal alteration and minerali-
zation in competent host rocks. In a complementary manner, pre-existing alteration such as
silicification may promote competency. Recent experimental evidence supports field observa-
tions that fracture permeability is promoted in extensional settings (Sibson, 1993; Cox, 1994).
Thus, magmatic fluid flow may be enhanced in competent host rocks adjacent to controlling
fault systems in settings such as hanging wall splits (e.g. Porgera Zone VII, Corbett et al., 1995),
or within extensional strike-slip systems (e.g. Nena, PNG, Corbett, 1994).

b) Styles

Differing styles of dilational environments are distinguished and in part display variations in-
dicative of the tectonic setting and levels of erosion of the hydrothermal system (Figs. 3.5,
3.6).

Styles of dilational ore systems include:

1. Splays
2. Tension fractures

32
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

3. Jogs
4. Hanging wall splits
5. Pull-apart basins
6. Pull-apart basin fracture arrays
7. Domes
8. Ore shoots
9. Pre-existing structures
10. Sheeted fractures

1. Splays or horsetail features develop along strike-slip structures as localising agents for
porphyry intrusion (e.g., Frieda Copper, PNG, Fig. 6.18; FSE, Philippines, Fig. 6. 24;
Chuquicamata, Chile). Splays which localise porphyry intrusions at depth probably propagate
upwards in dilational settings to form pull-apart basins in higher level epithermal environments.
Similarly, flower structures which are evident in profile in strike-slip or wrench faults display
pronounced vertical variations (Lowell, 1985). While positive flower structures form in
compressional settings, negative flower structures host basins.

2. Tension fracture/veins (McKinstry, 1948) form as dilational features within competent host
rocks between strike-slip faults, and are typically orientated in the direction of principal stress
(Figs. 3.9, 3.10), and commonly occur as en echelon arrays (Fig. 3.6). A variety of other terms
used are:
* tension gash is also valid for tension fractures, particularly where fractures display a
sigmoidal character (McClay, 1987; Figs. 3.6, 3.9),
* fissure is used by some workers to emphasise the steeply dipping nature of many
mineralized tension fractures,
* strike-slip duplex structures (Woodcock and Fisher, 1986) applies to similar features.

Tension veins are the most predominant form of dilational ore-hosting environment for epi-
thermal gold-silver vein systems. Examples are well developed in the Coromandel Peninsular of
New Zealand (e.g., Waihi, 5 M oz Au, Figs. 3.6, 8.6; Golden Cross, 1 M oz Au, Figs. 3.6, 8.2;
Thames goldfield New Zealand, 2 M oz Au; Figs. 7.44, 7.45). Others include Mt Kasi, Fiji (Fig.
6.28); Maniape, PNG (Fig. 7.36); the relationship of the 12 km long Nena Structural Corridor to
the controlling structures (Figs. 3.6, 6.19). These systems most strongly demon-strate the
manner in which regional strike-slip structures localise ore systems, but are them-selves
essentially barren (Figs. 3.5, 3.6).

Tension gash veins terminate along strike and so prospecting should focus on the identification
of new veins across strike (e.g., Waihi; Figs. 3.6, 8.6). Tension gash veins are initiated at
moderate angles to controlling structures and during mineralization rotate into higher angles to
the controlling structure. Thus, great care should be paid to the planning of the orientation of
drilling programmes.

3. logs form as bends in a throughgoing structure, which as dilational features, may host
mineralization and may therefore be distinguished from tension veins. While some deposits
clearly occur in jogs within a single structure (e.g., Karangahake, New Zealand; Cinola,
Canada, Corbett, unpubl. data), many are less obvious as discrete jogs (Cracow, eastern
Australia, Figs. 3.6, 8.7), and some are transitional to tension vein settings (e.g., Umuna Lode,
Misima, PNG, Fig. 3.6). A major jog in the Philippine Fault hosts the Southern Negros
geothermal field which is derived from actively intruding porphyry systems (Fig. 2.6).

4. Hanging wall splits form above dipping fault structures and are best propagated in
extensional settings characterised by normal faults. Many traditional models for hot spring -

33
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

style epithermal vein systems utilise hanging wall splits (Bonham, 1988). Hanging wall splits are
important loci of mineralization at Porgera (Fig. 7.23, Corbett et al., 1995) and Tolukuma (Fig.
7.48, Corbett et al., 1994c), and in both cases bonanza gold grades occur at the intersection of the
normal fault and hanging wall splits.

4. Pull-apart basins or grabens and half-grabens form as parallelogram-shaped features in

settings where the focus of movement crosses from one parallel strike-slip structure to another
(Fig. 3.6). Many pull-apart basins occur along the Sumatran Fault (Pudjowalujo, 1990), and are
associated with epithermal gold-silver vein systems (Kavalieris et al., 1987; Corbett unpubl
data). Pull-apart basins fill with epiclastic sediments which, as permeable host rocks, may
readily undergo alteration in settings where intrusives are emplaced into basins (e.g., Cinola,
Canada and others, Corbett, unpubl. data). Epithermal vein systems may form in dilational
environments within more competent basement host rocks adjacent to the pull-apart basins
(e.g., Way Linggo and Semung, South Sumatra, Indonesia, Corbett, unpubl. report, 1993).

5. Pull-apart basin fracture arrays are fracture patterns recognised in several vein systems.
These are characterised by steeply-dipping tension veins formed by rotation on controlling
structures and flatly dipping veins, which may have been initiated as extensile fractures
(Ramsay and Huber, 1987, p. 563), and enhanced within the extensional environment (Fig.
3.6). The steeply dipping tension veins may locally exhibit higher gold grades and act as fluid
flow feeders for the flatly dipping veins which tend to contain lower grade ores (e.g., Busai,
Woodlark, PNG, Figs. 7.32, 7.33, Corbett et al., 1994a; Ohui, New Zealand, Corbett, unpubl.
report, 1995). Similarly, gold mineralization at Lake Cowal, eastern Australia (2.44 M oz Au),
occurs in a setting of strike-slip deformation adjacent to the Gilmore Suture, and is best
developed in flatly dipping dilational quartz-carbonate-sulphide filled veins adjacent to steeply
dipping structures (North Limited, 1995). At Waihi, New Zealand (Fig. 8.6), the Empire vein
dips more shallowly than the Martha vein, and is locally parallel to a normal fault. The Antamok
vein system, Baguio, Philippines (10 M oz Au) displays a configuration of NW trending steeply
dipping higher grade veins and intervening lower grade shallowly dipping veins (Sawkins et al.,
1979; Damasco and Guzman, 1977). The NW veins are inferred to have developed by sinistral
rotation on controlling NS trending structures, which are parallel to the Philippine Fault, and act
as feeder structures to the intervening veins (Fig.
2.19).

7. Domes formed in basement rocks are inferred by some workers to be favourable settings for
mineralized vein systems (Mitchell and Carlile, 1994). The Hishikari vein system is aligned
along the intersection of a throughgoing linear identifiable for some distance on remote sensing
imagery (Corbett, unpubl. data, 1987) with a dome in basement Shimanto Group shales.
Similarly, Mitchell and Carlile (1994) infer a dome to localise the Acupan (4 M oz Au) and
Antamok (10 M oz Au) vein systems in the Baguio District, Philippines (Fig. 2.19). While each
are constrained between strike-slip structures, Antamok represents a pull-apart basin fracture
array, and Acupan occurs as en echelon tension gash veins aligned along a regional structure,
similar to Hishikari. The presence of a shoulder of barren high sulphidation alteration at Baguio
(Fig. 2.19, UNDP, 1987; Mitchell and Leach, 1991) and a rhyolite of roughly the same age as
mineralization at Hishikari (Izawa et al., 1990; section 8.vii.c) are each indicative of porphyry
intrusives at depth below the domes.

8. Ore shoots commonly develop as zones of increased vein width and gold grade, formed by
localised increased dilation within vein systems. Ore shoots occur as:
* Intersections of veins with diatreme breccias localise high grade ores (e.g., the GW
breccias at Acupan, Philippines, Sawkins et al., 1979; Tolukuma, PNG, Fig. 7.50,

34
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Corbett et al. 1994c; possibly Mangani, Indonesia, Kavalieris et al., 1987). These are
sites of increased mixing of fluids derived from magmatic source rocks and travelling
along the fractured diatreme margins.
* Shoots which form along dilatant vein/fracture systems at the intersections of cross
structures represent sites of higher fluid flow (e.g., Karangahake, New Zealand;
Cracow eastern Australia, section 8.vii.c.2; Thames goldfield, New Zealand, Section
7. iv.d; Bilimoia, PNG, Section 7.ii.d).
* Small scale jogs in linear veins.
* Rotations of vein segments constrained between domino faults (below, Fig. 3.9).

9. Pre-existing structures may be reactivated as dilatant ore hosts by changes in the stress
regime, typically in mesothermal vein systems adjacent to porphyry source rocks (e.g.,
Bilimoia and Arakompa, PNG, Section 7.ii.d; Tolukuma, PNG, Section 7.iv.d.3; Batu Hijau,
Indonesia).

10. Sheeted fractures form in porphyry and porphyry-related breccia environments in which
they may be dilated to represent favourable ore hosts described below.

vi) Fracture systems

Many dilatant ore-hosting structural environments display a history of activity extending from
pre- to syn-mineralization and commonly display some post-mineralization deformation of the
ore system. These structures may be reviewed in the light of fracture models. Several workers
have considered the original Riedel Shear Model (Riedel, 1929; Tchalenko, 1970; Tchalenko and
Ambraseys, 1970; Wilcox et al., 1973; Bles and Feuga, 1986; and many ot h-ers). Great care
should be exercised in the application to exploration examples of the Riedel Shear Model, which
was developed from experimental work and applies to fractures developed in sedimentary cover
overlying a strike slip fault in crystalline basement rocks (Fig. 3.7).

The Riedel Shear Model defines a set of specific structures which develop during strike slip
deformation of this type (Fig. 3.8). Experimental data indicates that the Riedel shears (R or
synthetic shears) form first and take up any early displacement. Conjugate (R1 or antithetic)
shears form later and display only minor displacements opposite to the overall rotation. The P
shears form last but take up much of the overall final shear displacement, especially in major
fault zones. Tension gash veins are initiated at higher angles to the R shears (45°) and are pro-
gressively rotated and dilated by rotation on the shear zone (Fig. 3.9). Continuing movement on
the controlling structures rotates the tension gash through the orientation in which dilation takes
place to a compressional orientation, whereupon new gash structures will be initiated (Fig. 3.9).
Tchalenko and Ambraseys (1970) recognised a similarity between the Riedel exper-iments and
fracture patterns developed in association with the modern earthquake at Dasht-e Bayaz (Fig.
3.4).

The Riedel shear model is most relevant to Pacific rim exploration in the manner in which ten-
sion gash veins develop, commonly in en echelon arrays. These are intimately related to the ore
hosting dilational environments in epithermal vein systems described above. For instance the
Golden Cross mine, New Zealand occurs as an en echelon vein array constrained between NS
structures (Fig. 8.2). The Taranaki-Hippo and Golden Cross vein systems were mined in the
Nineteenth century while the Empire vein is obscured by the Omahia andesite, and so was
identified more recently. This vein is currently being mined as a resource 0.9 M oz Au (Sec-tion
8.vii.c.2). Tension gash veins always display angular relationships to strike -slip

35
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Fig. 3.7

Fig. 3.8
Fig. 3.9
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

structures typical of dilation and are so distinguished from compressional fractures (Figs. 3.5,
3.9).

The fracture elements of the Riedel shear model may also be described in terms of a simple
compression model (Fig. 3.10). During compression, conjugate fractures are oriented at angles
<45° either side of the principle stress. Angles in the order of 60° between conjugate fractures
are common (Hobbs et al., 1976), in strike-slip settings (Bles and Feuga, 1986). Tension frac-
tures form parallel to the orientation of compression and bisect the conjugate fractures and in
these settings may not display the sigmoidal shape typical of tension gash veins (Fig. 3.10).
During compression or extension, rotation on conjugate structures may result in the develop-
ment to tension gash veins or jogs within the conjugate fractures, which may represent higher
grade dilatant portions of vein systems (Fig. 3.10).

As pointed out earlier, there is a growing body of evidence that porphyry intrusion takes place
during a relaxation of compression (extension) in magmatic arcs. Pre-existing conjugate frac-
tures commonly host mesothermal vein mineralization formed peripheral to porphyry intru-
sives. Higher grade zones develop in jogs within conjugate or tension gash veins formed be-
tween conjugate structures, and their shape is indicative of the orientation of rotation on the
conjugate fractures (Fig. 3.10). The orientations of dilational portions of conjugate vein sys-
tems adjacent to the Batu Hijau porphyry in the Banda Arc, Indonesia (Meldrum et al., 1994)
are indicative of intrusion during an extensional regime. Similarly, at the Grasberg porphyry
copper-gold deposit in the same setting, quartz stockwork veins conform to orientations con-
sistent with the activation of major local structures, including conjugate fractures (Kavalieris,
1994; Corbett, personal observation with G. MacDonald, 1994). Fracture orientations at the
Frieda Porphyry are also indicative of formation during extension and in an opposite stress re-
gime to pre- and post-mineralization compression (Asami and Britten, 1980).

Tension fractures aligned within the orientation of compression, commonly as reactivated pre-
mineralization structures, may host copper-gold mesothermal vein mineralization proximal to
porphyry environments (e.g., Arakompa, PNG, Fig. 7.10).

Empirical observations of many Pacific rim vein systems (Corbett, unpubl. data.) suggests that a
set of cross structures may conform to a model in which these structures both influence ore
formation and act as post-mineralization offsets. The parallelism has caused them to be termed
domino faults. The orientation of these structures is inconsistent with either a P or R' shear in the
Riedel Model. Rather, they coincide (Fig. 3.8) with the position in which folds or X fractures of
Logan et al., (1979) and Swanson (1988), which are initiated as conjugate frac-tures during
layer-parallel extension. These structures are inferred to have formed early dur-ing extension and
may therefore have been activated during continuing deformation associated with mineralization.
Fluid upflow features have been recognised at the intersections of domino faults and dilatant
tension gash structures (Fig. 3.9). Domino faults are more commonly noted as providing post-
mineralisation offsets to vein systems during strike-slip deformation, by act-ing as block faults to
facilitate shortening and so offsets. Similar fractures are noted to facili-tate block rotations
between faults at the Mesquite Mining District California (Willis and Tosdal, 1992) and
elsewhere by Hanmer and Passchier (1991). Rotation on the domino faults is opposite to that of
the bounding strike-slip structures. Activation of the domino faults dur-ing the mineralization in
epithermal vein systems may cause enhanced dilation in vein seg-ments, constrained between
these structures to form ore shoots (Fig. 3.9).

The same domino cross structures:

* localise fluid upflow zones during extension, typically in magma fluid dominated

36
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralisation" Corbett G J & Leach T M, 8/96 Edn.

Fig. 3.10

Fig. 3.11
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

systems (e.g., Mt Kasi, Fiji),

* form high grade shoots in banded vein systems (e.g., Semung, Indonesia),
* act as post-mineral vein offsets (e.g., Tolukuma, PNG).

vii) Shear sense indicators

An understanding of the sense of displacements of faults may be an important exploration tool

at different scales.

On a regional scale, the displacement on many major structures may be predicted from the tec-
tonic setting. For example, structures parallel to the Philippine, Sumatran, or San Andreas Faults
are likely to display rotations consistent with those major structures, as governed by plate
movements (Fig. 3.1). Thus, in terrains dominated by strike-slip structures of consistent
orientation, dilatant subsidiary structures should also display consistent orientation. The
McLaughlin gold deposit is localised by strike-slip movement on structures parallel to the San
Andreas Fault (Tosdal et al., 1993; Donnelly-Nolan et al., 1993). A prospecting tool may emerge
from an understanding of the regional structure. If the direction of movement on a co n-trolling
regional structure is known, it might then be possible from the orientation of subsidi-ary
structures to distinguish which structures are more likely to be dilatant and mineralized from
those which are in compressional orientations and hence unmineralized (Fig. 3.5). This procedure
assumes consistent senses of rotation.

In the Coromandel Peninsula, New Zealand, it has long been recognised that most vein sys-tems
trend northeast (Christie and Brathwaite, 1986). This is caused by dextral rotation on the
Coromandel Peninsula derived from movement of the Pacific plate against the Australian plate
(Fig. 7.44). Fissure vein systems such as Martha Hill, Golden Cross, Tui, and Karangahake are
hosted within dilatant subsidiary structures.

At the prospect scale, shear sense indicators are also discernible from the application of the
Riedel Shear Model to fracture systems recognised during field mapping. The orientation of
dilatant versus compressional subsidiary structures and domino faults may provide sense of
movement indicators on major structures, and assist in the exploration for faulted-off vein sys-
tems. Recognition of the direction of movement on weakly mineralized major structures may aid
in the identification of the orientations of dilational subsidiary fractures and assist in the planning
of trenching or drilling programmes (Fig. 3.5). High-grade ore shoots formed within a domino
fault model may also represent important exploration targets.

Outcrop scale sense of shear indicators may be indispensable in the determination of offsets of
vein-style ore bodies, particularly in underground mines. Slickensides, scratch marks, ploughed
striations, or grooves and crystal growths, are indicators of whether a fault displays dip-slip or
strike-slip rotation. However, the sense of which block is up/down or left/right as defined by
irregularities in the slickensides is more difficult. Petit (1987) categorised secondary fractures in
terms of the Riedel shear model as P, R or T shears to provide direction of movement indi-cators.
These features may be difficult to identify in outcrop. Plough marks or mineral growth fibres
provide easily discernible indicators of the direction of movement (Fig. 3.11). The "smooth-
rough" rule suggests that when a hand is run over fibres within a fault, the direction which feels
smooth, is that of block movement (Mawer, 1992).

37
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

viii) Porphyry- and intrusion-related fracture patterns

In compressional arcs porphyry intrusions are inferred to have commonly been emplaced dur-ing
a relaxation of compression (extension). Similarly, in settings of oblique subduction, porphyry
intrusions are localised at dilatant features (splays or jogs) within accretionary struc-tures, or at
the intersections with cross structures such as transfer structures. Many porphyry gold-copper
ore bodies are inferred to have formed as apophyses to larger magma sources at depth from
which much of the mineralization has been bled (Section 5.c). The low and high sulphidation
gold-copper mineralization discussed herein is derived from magmatic source rocks, but occurs
outside the porphyry environment. Fracture systems are therefore important in fluid transport
from the magma source to sites of deposition, and mineral deposition.

A common ore hosting environment in porphyry-related Pacific rim mineral systems is sheeted
fractures formed in association with porphyry intrusion which become dilated by syn-mineral
deformation. The integration of studies of fracture patterns resulting from intrusion emplace-ment
(Phillips, 1974; Koide and Bhattacharji, 1975) with those of dilational structural environ-ments
(above) may aid in the evolution of models for ore deposition.

Fracture systems associated with porphyry intrusion occur as:

1. Reactivation of existing fracture/breccia systems,

2. Radial fractures, commonly exploited by radial dykes,
3. Ring fractures and dykes,
4. Sheeted fractures,

1. Existing fractures and breccias are common mediums of fluid transport in porphyry systems.
These may have been reactivated as dilational features during porphyry intrusion and the
subsequent evolution of fluids, and may reflect the dilational structural environment into which
the porphyry was emplaced. Many mineralized porphyry copper-gold deposits are inferred to
represent cylindrical-shaped apophyses to larger magma sources at depth (e.g., Goonumbla,
eastern Australia, Heithersay et al., 1990; Grasberg, Indonesia, MacDonald and Arnold, 1994;
Philippine deposits, Sillitoe and Gappe, 1984). The fractured and brecciated contacts to the
apophyses are common mediums of fluid transport. Stockwork quartz veins formed during
porphyry emplacement may be cut by later fractures which transport fluids and metals.

2. Radial fractures formed by the initial porphyry intrusion are commonly exploited by dykes
(Fig. 3.12). The radial pattern may be modified by any pre-existing structural grain and can
extend for some distance from the porphyry source. The fracture pattern of mineralization at
Grasberg is indicative of the enhancement of regional throughgoing fractures as conjugate
joints and the exploitation of these by mineralized fluids (Kavalieris, 1994).

3. Ring dykes commonly form as late stage intrusives which exploit concentric fractures
formed by magma contraction in caldera ring fracture settings. In settings such as Emperor gold
mine (Eaton and Setterfield, 1993) ring fractures form hosts for mineralization developed in
distal settings to the porphyry source, especially at the intersections with cross structures.

4. Sheeted fractures and veins are an important mechanism of fluid transport in many porphyry
copper-gold and porphyry-related breccia gold systems. The term sheeted fractures has been
used in the geological literature to describe parallel fractures which form as flat lying extension
joints in granitic rocks (Price, 1966; Hobbs et al., 1976). These probably form as a

38
Exploration Workshop "SW Pacific Him Au/Cu Systems: Structure Alteration & Mineralisation" Corbett G J & Leach T M, 8/96 Edn.

Fig. 3.12

Fig. 3.13
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

result of erosion-induced stress release, particularly in glaciated terrains (Price, 1966). More
dynamic stress inducted structures are of interest to the exploration geologist in porphyry ter-
rains. These conditions may give rise to steeply dipping sheeted fracture/veins which overlie
porphyry systems as important ore hosts.

Sheeted fracture/veins

Many mineralized porphyry intrusives occur as cylindrical-shaped apophyses to larger magma

sources (e.g., Goonumbla, eastern Australia, Heithersay et al., 1990; Grasberg, Indonesia,
MacDonald and Arnold, 1994; Philippine deposits, Sillitoe and Gappe, 1984). The forceful ver-
tical emplacement of the cylindrical intrusive causes cone-shaped sheeted fractures to form in the
overlying rocks (Phillips, 1974). Cone-shaped sheeted fractures ring the top of the intru-sive as
kinked straight segments, dipping moderately inward close to the intrusive and steepen-ing at
higher levels and further from the intrusive (Fig. 3.14). There is a small element of hori-zontal
stress near the intrusive but only a vertical component at higher levels. Concentric frac-tures
propagate above intrusions with high interstitial fluid pressures compared to lithostatic pressures
and form concave upward shapes (Koide and Bhattacharji, 1975). Stockwork frac-tures form in
the carapace and sheeted fractures predominate above the intrusive margins. These are well
placed to act as plumbing systems for fluids which migrate along the margins of the intrusive and
into the overlying sheeted fractures (especially where dilated, see below). Overprinting intrusives
and the vertical nature of the apophyses promote the development of sheeted fractures within
earlier intrusives, which become hosts for later mineralization, derived from intrusions at depth.
Large-scale sheeted fractures accommodate collapse as caldera ring fractures. Others act as a
locus of intrusion of breccia systems and so may form funnel-shaped breccia pipes.

The term stockwork breccia/fracture/veins is distinguished from sheeted fractures and used to
describe more randomly oriented network veins such as the quartz veins which characterise
porphyry systems. Although the overall geometry is similar, the presence of veins distin-guishes
stockwork veins from crackle breccias (below). The random stockwork veins form in settings
where stress regimes are not pronounced, typically in the upper portions of porphyry intrusions,
whereas sheeted veins are inferred to form overlying porphyry margins (Fig. 3.12), or marginal
to subvolcanic breccia pipes (Fig. 3.15).

Role of "retrograde boiling"

"Retrograde boiling" is probably better termed the exsolution of volatiles due to crystallization of
a melt (Shinohara and Kazahaya, 1995). This is an important process in the formation of copper-
gold mineralization within the porphyry environment as well as in fracture and breccia systems
above the source porphyry (Phillips 1973, 1986; Burnham, 1985). A detailed discus-sion of
retrograde boiling is given by Burnham (1979) and summarised by Pirajno (1992). Many
mineralised porphyry systems occur as apophyses into which volatiles have collected from larger
magma bodies at depth. Cylindrical shaped intrusions are noted by many workers (e.g.
Philippines, Sillitoe and Gappe, 1984; southwest USA, Lowell and Gilbert, 1970; Gras-berg,
Indonesia, MacDonald and Arnold, 1994; Goonumbla, eastern Australia, Heithersay et al., 1990).
Volatile-rich intrusions display lower viscosities and so are emplaced more easily to higher
crustal levels, as apophyses. Intrusions form a carapace comprising the chilled margin surrounded
by thermally metamorphosed country rock and then progressively cool inwards. The marked
reduction in confining pressure, and hence solubility, as a magma emplacement to a high crustal
level promotes an exsolution of volatiles. This is termed first boiling by P irajno (1992).

39
 Exploration Workshop 'Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

The exsolution of volatiles results in a volume increase as the cooling intrusive separates into
crystal and the volatile components. This may be accentuated by the collection of volatiles in
the apophyses to a larger body of cooling magma. Intrusives forcibly emplaced to high crustal
levels will cool more quickly. At these levels the lithostatic confining pressure is lowered, but
partly compensated for by the increased tensile strength of the carapace of the inward cooling
magma. The carapace essentially seals the system and encloses the saturated and overpressured
fluids. Traditional models (Phillips, 1973; Burnham, 1979) focus upon fracture of the carapace
at the point at which the vapour pressure exceeds load pressure + surface tension + tensile rock
strength. Disruption by faults which control porphyry emplacement in tectonically active
magmatic arcs may fracture the carapace before that point is reached. The disruption of the
overpressured carapace dramatically lowers the vapour pressure, prompting additional exsolu-
tion of volatiles. The pressurised fluids may escape as:
* explosive breccias,
* promote hydraulic fracturing of the carapace,
* sheeted fractures.

Explosive breccias are discussed below. Rocks which have undergone hydraulic fracturing
provide a large surface area for alteration by the venting hydrothermal fluid, commonly as in-
jection breccias (below). Quenching in response to rapid changes in conditions may promote
metal deposition (e.g., high gold grades breccias at Lihir, PNG, Moyle et al., 1990). The mar-
gins of cylindrical intrusions are overlain by conical sheeted fractures which propagate up-
wards. The rapid depressurisation dramatically lowers the solubility of quartz (Fig. 4.2) which
infills fractures as veins.

Dilated sheeted fractures

Many southwest Pacific rim porphyry systems are emplaced into active plate margins and
localised along regional accretionary (arc-parallel), strike-slip structures by jogs, splays, or
intersections with transfer (arc-normal) structures (Fig. 3.13). The stress environment active
during initial intrusion commonly continues during sheeted fracture formation, degassing of the
intrusive, and mineralization. In environments of oblique collision, and associated strike-slip
deformation on regional controlling structures, the sheeted fractures aligned along the ten-sion
gash orientation become dilated and mineralized while others undergo compression (Fig. 3.13).
These preferentially dilated sheeted fractures which overlie intrusives represent a medi-um
which may host mesothermal vein mineralization. Many porphyry systems display differ-ing
intensity of sheeted fracture development in relation to the regional stress regimes. Dila-tant
structures not only focus the upward moving magmatic fluids but also host ground wa-ters, to
promote mineralization by mixing.

In environments of orthogonal compression, radial fractures in the conjugate orientations tend

to be dilated and mineralized. Rotation on these fractures may produce higher metal grades in
sigmoidal dilatant zones (Figs. 3.10, 3.13). However, significantly more fluids tend to be bled
from the porphyry source in settings of oblique rather than orthogonal compression.

40
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

ix) BRECCIAS

a) Introduction

Practically all magmatic arc copper-gold systems discussed herein contain breccias, and so the
processes of breccia and ore formation are intimately related. As exploration geologists we seek
to further understand the relationship to mineralization of the tremendous variety of brec-cia
types. For instance, the mapping of barren diatreme breccias may point towards the miner-
alization elsewhere in the hydrothermal system. Yet the terms used are as numerous as the ge-
ologists involved. The aim here is to:
* focus on the processes of formation,
* to delineate differing breccia types, and
* describe their role in ore formation.

We attempt to maintain consistency with existing breccia classifications and terminology, a lbe-it
in the light of our own personal experience. Readers are referred to the following works, on
which this discussion draws, for more detailed and in some instances alternative analyses of
breccias; Sillitoe (1985), Baker et al. (1986), and Taylor and Pollard (1993).

What is a breccia?

A breccia is a clastic rock composed of fragments held together by matrix and containing
cavities filled by post-brecciation hydrothermal minerals (Taylor and Pollard, 1993).

Fragments or broken rock clasts become progressively milled with increased deformation
(brecciation). Some breccias contain only host-rock fragments while others are characterised
by introduced fragments, and fragments undergo varying degrees of alteration.

Matrix is the fine rock material between the fragments and, depending on the degree of mill-
ing, may be gradational to the fragments. Breccias are either matrix or fragment (clast) sup-
ported. The majority of the mineralized component of breccias is introduced as hydrothermal
fluid and so occurs within the matrix.

Cavities develop during breccia formation and are infilled with hydrothermal minerals in-
cluding gold-copper mineralization, and so are an integral part of the brecciation and min-
eralization processes.

All these fragment and matrix variations contribute towards the development of differing
breccia types.

b) Classification

We need to name (classify) breccias in order to map out hydrothermal systems. Breccias may
be distinguished by:
* appearance or a descriptive classification,
* mode of formation or a genetic classification.

Which is appropriate?

When we study a breccia it is advisable to start by recognising the differing breccia types or
textural variations and map them out using descriptive terms.

41
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Descriptive terms are most useful in order to provide the data to map out a breccia at the ini-tial
stages of an investigation. However, breccias of similar appearance may be derived from
different processes. Thus as the knowledge of a particular system evolves, an understanding of
the context may allow a breccia to be put in a more genetic framework. A mosaic breccia, which
is described simply as, "one in which the fragments can be fitted back together by re-moval of
the matrix", can be derived by a variety of breccia processes. A mosaic breccia in the carapace
to an intrusion may be barren and not vector towards higher grade mineralization in the same
manner as a mosaic breccia in a high-sulphidation system.

As the descriptive delineation of a large body of breccia emerges, it may then be possible to
determine the geological environment in which the breccia formed and the process of for-
mation. A genetic classification for the breccia might then emerge. Knowing the type of brec-cia
could hasten the analysis that particular style of system.

A genetic term is based on the interpreted manner of formation of a breccia and as such is de-
rived from only the data to hand a that stage. Recording information in using a genetic termi-
nology may result in a loss of the original data-base and inhibits later reinterpretation, espe-
cially as the knowledge of the system evolves, such as by the progression from mapping to di-
amond drilling. The genetic term "diatreme breccia" might be used to describe a breccia formed
in association with a maar volcano/diatreme. The descriptive terms "milled matrix flu-idised
breccia or polymictic muddy breccia" could describe the same rock. There is an o bvi-ous
tendency to drift towards the shorter and easier genetic terms. However, the milled mud-dy etc.
breccia is noi a diatreme breccia until it has been put in the context of the entire mapped maar
volcano/diatreme breccia system.

As a general rule - it important to avoid early genetic descriptions. Decriptive breccia terms

Some descriptive breccia terms are useful, although breccias described in this manner, may
form within a variety of genetic settings.

Dilational breccias form by the infilling of open space by later fluid, commonly in brittle rocks
in high level settings. A colloform/crustiform banded breccia matrix may indicate re-peated
fracture opening and rapid infilling by quenched hydrothermal fluid, whereas slower cooling
will result in banded crystalline infill. Rebrecciation and cross-cutting infill are char-acteristic
of these breccias. This breccia term is indicative of a dilational ore-forming process.

Mosaic or jigsaw breccias are classified as those in which the fragments may be joined back
together by removal of the matrix. Thus, these breccias exhibit no input of exotic fragments
and little fragment rotation or rounding. The matrix tends not to be of locally derived milled
material, but composed substantially of introduced hydrothermal components. There is obvi-
ously an overlap between the use of the terms dilational and mosaic breccias.

Milled matrix fluidised breccias encompass a variety of typically phreatomagmatic breccias

composed of milled fragments in which the matrix is made up of comminuted fragment ma-
terial. This descriptive term applies to breccias which might be also described by a genetic
term as belonging to diatreme/maar volcano complexes. Many exploit structures as dyke-like
forms and have been termed tuffisite by some workers (Paull et al., 1990).

42
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbert G J & Leach T M, 8/96 Edn.

Genetic Groupings

A breccia classification based upon the environment of formation distinguishes groups of

breccias according to the relationship with a magma source and partly those formed at dif-
fering crustal levels. These as partly illustrated in Fig. 3.14 and classified as:

i) Primary non-hydrothermal breccias.

* Magmatic breccias
* Volcanic breccias
* Tectonic breccias

ii) Ore-related hydrothermal breccias.

* Magmatic hydrothermal breccias
* Phreatomagmatic breccias
* Phreatic breccias

c) Primary non-hydrothermal breccias

Magmatic breccias form in association with porphyry intrusions in which there is little mixing
of meteoric waters or fluid degassing. Only with the input of hydrothermal fluid are these
breccias inferred to become mineralized. Intrusive or contact breccias develop at the contacts
between intrusions and host rocks, while collapse breccias form following the outflow of ma-
terial from a magma chamber.

Volcanic breccias are the large group of broken rocks which form in subaerial and subvolcanic
environments. Although volcanogenic massive sulphide deposits form in association with these
breccias, the distal relationship to porphyry source rocks excludes them from this discussion.
Readers are referred to texts such as Cas and Wright (1987) and McPhie et al., (1993) for a
discussion of these rock types.

Tectonic breccias develop by deformational processes and include fault breccias which vary
from milled puggy fault zones to open space breccias and are usually distinguished by a rela-
tionship with planar fault surfaces. However, many faults are plumbing systems for hydro-
thermal fluids as so these breccias are transitional to hydrothermal phenomenon.

c) Ore-related hydrothermal breccias

These are grouped as.

1. Magmatic hydrothermal breccias
2. Phreatomagmatic breccias
3. Phreatic breccias

These three classes of breccias are distinguished using the terminology of Sillitoe (1985) on the
basis of crustal level and relationship to porphyry source rocks (Fig. 3.14). Magmatic breccias
typically form at deepest or porphyry levels and are eroded to display pipe -like forms, but need
not have vented to the surface. Phreatomagmatic breccias typically display associations with
high level porphyry intrusions and may vent as diatreme/maar volcanos or remain as milled
matrix fluidised breccias. Phreatic breccias form a surficial levels and in this classification do not
display a relationship with high level intrusions. Differing styles of low sulphidation gold
deposits display associations with varying breccia types developed at differ-ent crustal levels
(Figs. 7.1, 7.2).

43
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

1. Magmatic hydrothermal breccias

Magmatic hydrothermal breccias (Figs. 3.14, 3.15) are characterised by a pronounced magmat-ic
involvement in the brecciation process and a hydrothermal ore fluid dominated by magmatic
component (e.g., in eastern Australia, Kidston, Baker and Andrew, 1991; and Mt Leyshon, Paull
et al., 1990; in Chile, San Cristobal, Corbett, unpubl. reports, Egert and Kasaneva,
1995). The Kidston breccia is described here and within the case study of the associated min-
eralization (Section 7.ii.d.v).

Subvolcanic breccia pipes, which host hydrothermal magmatic breccias typically form at con-
siderable depths (>1 km) equivalent to high level, commonly felsic, porphyry intrusions, and are
generally not expected to have vented to the surface. Breccia pipes commonly overlie apophyses
to larger bodies of magma, from which metals at a low tenor may have been con-centrated.
Thus, as in the case of Kidston, the analyses during exploration of the architecture of buried
intrusive source, may assist in the identification of settings for breccia pipes, typical-ly at cross
structures (Fig. 7.6). Similarly, the Mt Leyshon gold breccia is localised by the in-tersection of a
regional structural corridor with the margin of a subvolcanic complex (Paull et al., 1990).

The mechanism inferred for breccia formation relies upon the violent explosion of volatiles
during retrograde boiling described above. Volatiles, possibly derived from a large magma
source at depth, collect and become overpressured in apophyses as intrusions cool. Movement
on a controlling structure may fracture the carapace allowing the volatiles to vent explosively.

It is important to stress that most breccia pipes represent pre-mineral explosive venting of
volatiles and that metals subsequently exsolve from a deeper parent magma. The breccia for-
mation taps the top of the magma chamber and fractures the overlying country rocks to pro-vide
a focus for the degassing fluids. Mineralization partly fills open space within breccias, and
(sheeted) fractures, which are best placed as sites for exsolving metals. An understanding of the
anatomy of breccia pipes may provide vectors towards mineralization. Baker et al., (1986)
distinguish different levels of breccia pipes and Sillitoe (1985) describes upper termina-tions
into collapse breccias and the rarely seen lower contacts as; fissures, source intrusions, or shears.
Lateral contacts are commonly sharp and exhibit sheeted fracturing (Sillitoe, 1985).

Sheeted fractures are inferred to have developed during the initial explosive venting related to
retrograde boiling and have no doubt been active during later collapse. They dip steeply and
are kinked about pipe margins (e.g., Kidston, Fig. 7.7; Cabeza de Vaca, Chile, Sillitoe and
Sawkins, 1971).

Two principle facies within breccia pipes related to the mechanism of formation are:
* intrusion breccias
* collapse breccias

The genetic term intrusion breccias describes rocks developed in conditions of intense fluidi-
sation during the explosive venting of volatiles injected from the intrusive into the overlying
host rocks, and forms breccias which include those described as milled (Baker et al., 1986), or
rock flour (Sillitoe, 1985) breccias. These breccias are generally supported by a matrix of
comminuted rock flour and hydrothermal cement, and represent mixes of rounded, commonly
competent, intrusive fragments which have been milled during transport. They grade laterally
into breccias dominated by subangular locally derived fragments and jigsaw

44
Exploration Wciksitcp *SW Pacf;s K_rr> AiiiCu Systems: Structure Alteration & Mineralization' Corbett G J & Leach T M, 8/96 Edn.

Fig. 3.14

Fig. 3.15
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Ecln.

breccias, of the collapse type. Milling takes place in highly fluidised environments. These brec-
cias form in the core of the breccia pipe (Fig. 3.15). Early mapping at Kidston (Corbett, un-publ.
data) defined a zone dominated by intrusive fragments (termed volcanic breccia in Fig.
7.7), which roughly corresponds to pre-breccia rhyolite early tourmaline breccias of Baker and
Andrew (1991). Other breccias of this type rim intrusive dykes (e.g., San Cristobal, Chile). At
both Kidston and San Cristobal, early fine grained felsic dykes associated with the intrusive
breccias are cut by later coarser grained quartz feldspar porphyry intrusives, possibly reflecting a
melt derived from deeper within the magma chamber.

Pebble dykes comprise linear bodies of well rounded, commonly transported fragments in a
milled matrix formed by the venting of volatiles in faults or joints. They are common in porphyry
environments (e.g., El Salvador, Gustafson and Hunt, 1975), and locally provide ground
preparation for mesothermal vein mineralization (e.g., Arakompa, PNG, Corbett et al., 1994b).
Sillitoe (1985) however, cites the association with specific intrusive phases to suggest that pebble
dykes are more commonly late- to post-mineral in age. Pebble dykes contrast with fluidised
breccias (below) which are characterised by transported matrix.

The genetic term collapse breccias is applied to a group of open space breccias formed during
relaxation following the initial explosive intrusion breccia event, and are typically best ex-posed
about the outer, particularly the upper, portions of breccia pipes. Only small degrees of transport
of country rocks are recognised in many typically jigsaw (Sillitoe, 1985) or shatter (Baker et al.,
1986) breccias which can be joined back together removing the hydrothermal cement, and
contrast with the introduced fragment of the intrusion breccias. A country rock geological
contact can be traced from outside to within the Kidston breccia pipe and the pres-ence of large
blocks are indicative of a transition to a margin of the system (Fig. 7.7). Sheeted fractures
formed during initial explosion may facilitate collapse.

The terms shingle, domino. (Sillitoe, 1985) or imbricate (Baker et al., 1986) apply to slab-like
breccias formed by regular breakage or sheeting about the margins of breccia pipes. While
these breccias may dip flatly above a pipe, Sillitoe (1985) emphasises the shallowing in dip
moving inward from the pipe margin. These are transitional to the angular shatter breccias of
Baker et al., (1986).

Decompression breccias form during the rapid depressurisation of venting fluids or rock mass-
es. They exhibit the appearance of spheroidally weathered and rounded fragments, rimmed by
curved and tabular (Baker et al., 1986) fragments, separated by minor open space which may be
infilled with hydrothermal minerals (Fig. 3.14). These breccias may be likened to the hypo-gene
exfoliation described by Sillitoe (1985).

Magmatic hydrothermal injection breccias are distinguished as those breccias derived from a
magmatic source, but composed of only hydrothermal fluid, hence the name applies to "hot
water" breccias. The terminology for hydrothermal breccias parallels with the term hydraulic
breccia used in the literature to describe breccias formed by breaking under the influence of
pressurised hydrothermal fluids. Many of these breccias form in subsidiary structural envi-
ronments derived by the rotation of major structures (section 3.iv), and so display transitional
relationships to the dilational breccias. Hydrothermal breccia styles may vary according to the
degree of mineralized fluid input and therefore display a relationship with metal grades (Figs.
3.16, 6.3). Rotational breccias which are characterised by substantial fragment rotation or
transport in association with considerable fluid injection, display the highest metal grades.
These pass with smaller quantities of injected fluid to mosaic or jigsaw breccias, characterised
by fragments which are separated but have not undergone substantial transport. Fluidised brec-
cias are distinguished on small scales as containing milled fragments

45
Exploration Workshop 'SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralization" Corbett G J & Leach T M, 8/96 Edn

Fig. 3.16

Fig. 3.17
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

in a transported and exotic matrix, and commonly exploit fractures, to form dyke-like bodies.
Where only minor matrix introduction is recognised within fractures, then mosaic-like crackle
breccias are recognised. If crackle breccias are opened and exhibit fluid flow, then a fluidised
crackle breccia may result. Thus a pattern emerges where copper-gold grades may be directly
proportional to matrix content in breccias, which change with increasing distance from the
magmatic source from; rotational -> mosaic -> fluidised -> crackle breccias (Fig. 3.16). An
ability to map out zones of breccia types may provide vectors to higher grade ores or magmat-ic
source rocks, particularly in high sulphidation copper-gold systems (Fig. 6.3).

Hydrothermal collapse breccias form during the retrograde phases of porphyry copper devel-
opment. Magmatic vapour rises above the porphyry environment, condenses, mixing with
ground waters and collapses as hot low pH fluids (Fig. 5.7). Pressure draw-down during the
waning stages of cooling porphyry may assist in the fluid collapse. The resulting overprinting
phyllic to argillic alteration forms clay matrix breccias in which sericite and clays grade from
fractures or crackle breccias into the host rock. Remnants of original rock type which remain
generally display no rotation, so that the hypogene alteration results in a fabric in which cores
of unrotated primary rock are set in a clay altered matrix. Fluids commonly migrate down
structures which, as zones of incompetent clay alteration, may be reactivated. Thus, many clay
matrix breccias grade to shear zones which may display intense fragment milling in areas of
high strain. Hydrothermal collapse and associated retrograde phyllic and argillic alteration may
represent a mechanism for the upgrading of the mineral tenor in porphyry systems.

2. Phreatomagmatic breccias

Phreatomagmatic ("phreato" meaning water converted to steam and magma) eruptions are the
violent eruptions which may occur when groundwaters are superheated by contact with rising hot
magma and rapidly evolve into steam. A non-genetic term "milled matrix fluidised brec-cia"
could be used to describe breccias formed by this process. These breccias commonly ex-ploit
pre-existing structures which may also host volatile-rich intrusions. A venting eruption is
recognised at surficial levels as a maar volcano, and diatreme breccia complexes may extend to
considerable depths, where relationships with high-level porphyry intrusions become more
apparent. Thus, the term of diatreme breccias is applied to rocks which develop by phreato-
magmatic processes (Fig. 3.17). Although vapour driven, diatreme breccias display a strong as-
sociation with high level porphyries, commonly within flow dome complexes. Porphyry-related
mineralization is generally of the carbonate-base metal style for most low sulphidation systems
(Fig. 7.2), described as deeper epithermal by Sillitoe (1985), and diatreme breccias are also
common in high sulphidation systems (Fig. 6.1).

Examples of diatreme-related gold/copper mineralization in the southwest Pacific include: in the

Philippines at Acupan, 4 M oz Au (Damasco and Guzman, 1977; Sawkins et al., 1979; Cooke
and Bloom, 1990); Lepanto, >3 M oz Au (Section 6.iv; Garcia, 1991), and Dizon 3 M oz
(Sillitoe and Gappe, 1984); in Papua New Guinea at Wau (Sillitoe et al., 1984), Kerimenge,
1.8 M oz Au (Section 7.iii.j; Hutton et al., 1990), Edie Creek (Section 7. iii.j; Corbett, 1994),
Tolukuma (Section 7.iv.d.3; Corbett et al., 1994a), and Wafi (Section 6.iii.b; CRA, 1994); in
Indonesia, Kelian, >4 M oz Au (Section 7.iii.j; van Leeuwen et al., 1990, Sillitoe 1995); and
Miwah, (Williamson and Fleming, 1995); and Gold Ridge, Solomon Islands (Section 7.iii.j;
Sillitoe, 1989, Fig. 7). Others include: in the USA, Cripple Creek (Thompson et al., 1985),
Montana Tunnels (Sillitoe et al., 1985), and Pueblo Viejo, Dominican Republic (Vennemann et
al., 1993).

Diatreme breccia/maar volcano complexes form in relation to high level porphyry intrusion

46
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Fig. 3.18

Fig. 3.19
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

and so are inferred to have been generated at considerable depth to at least 1 km, as evidenced
by the depth of formation of associated carbonate-base metal mineralization. Diameters may
range to several hundred metres across and repeated activation results in rebrecciation and
overprinting diatremes. Flared margins are inferred to dip steeply inward at depth and shallow
closer to the surface to produce the overall funnel shape (Sillitoe, 1985).

Diatreme breccias are commonly localised by major structures along which high level porphyry
intrusives have been emplaced and which may have been reservoirs for groundwaters. Exam-ples
include: the graben structure at Tolukuma (Fig. 7.47, Corbett et al., 1994a); Escarpment Fault at
Wau, (Figs. 7.28, 7.29; Sillitoe et al., 1984); Lepanto Fault at Lepanto (Fig. 6.23; Baker, 1992).
Some diatremes are localised by cross structures (e.g., Tolukuma; Fig. 7.47), or by dilational
splays from more major structures (e.g., Lepanto; Fig. 6.23).

Breccias within diatreme complexes are characterised by milled matrix fluidised breccias. These
vary from well milled hetrolithic breccias comprising country rock and introduced porphyry
fragments, typically in the most activated regions of the diatreme complex, to more angular
monolithic breccias, typically towards the periphery. Most breccias are supported by a matrix of
comminuted rock material which generally displays clay pyrite alteration and so con-tain no
open space. Tuffisite rocks (Cloos, 1941 in Sillitoe, 1985) comprising well milled, in-tensely
altered, commonly bedded, tuffaceous material, are characteristic of diatreme breccias. These
may exhibit dyke-like forms and display a genetic relationship to mineralization (e.g., Mt
Leyshon, Eastern Australia, Paull et al., 1990).

The distinction between tuff ring and vent facies may be useful when mapping diatreme brec-
cias, as mineralization is commonly localised at diatreme margins (Fig. 3.17). Tuff ring facies or
tuff apron (Baker et al., 1986), comprises material ejected from the diatreme and deposited
outside the actual vent and is only preserved in systems which are poorly eroded. Various
workers (Sillitoe, 1985; Baker et al., 1986; Cas and Wright, 1987) describe base surge deposits
derived from the lateral movement of the rapidly expanding gas cloud. The resulting deposits
may form as thin laterally extensive layers containing exotic blocks and exhibit local low angle
cross stratification. Finer grained tuffisite layers may contain accretionary lapilli formed as the
gas cloud condenses and fine grained (mud) particles adhere to a nucleus. The recognition of
accretionary lapilli has formerly been taken by many workers as evidence for a surficial origin of
diatreme breccias. However, similar features have been generated in subsurface settings in
experimental conditions (McCallum, 1985). Accretionary lapilli are recognised in rocks formed
in subsurface environments at Nena, PNG (Leach and Corbett, pers. observation) and Mount
Leyshon, Australia (I Hodkinson, personal commun.). Sillitoe (1985) also stresses the collapse of
base surge deposits into diatremes such as at Cripple Creek (Thompson et al., 1985).

Vent facies comprise the main body of the diatreme vent. The maar volcano represents the
surficial portion of the vent and is commonly infilled with lacustrine sediments. Breccias are
locally rebrecciated and comprise introduced intrusive fragments, milled rock flour, tuffisite, and
blocks slid in from the sides commonly fill the vent (Fig. 3.18). Instances of considerable
collapse are noted by McCallum (1985) and Sillitoe (1985) and include; shale fragments trans-
ported 1500 m down into the Mule Ear diatreme Utah (Stuart-Alexander et al., 1972), charcoal
fragments occur 650 m below the present surface at the Balatoc diatreme (Sawkins et al.,
1979), from which probably some 400 m has been eroded, and the base surge deposits at >300 m
depth at Cripple Creek (Thompson et al., 1985). Well milled hard intrusive fragments re-flecting
considerable vertical transport may occur with breccias components dominated by angular,
softer, locally derived rocks. Fragment and matrix types reflect the host rock type and

47
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

the driving intrusive. The recognition of mineralized fragments within diatreme breccias may
provide an indication of a target at depth, as with diatreme breccias in the vicinity of Lepanto,
(Sillitoe, pers. commun.).

Dyke-like fluidised milled matrix breccias breccias represent non-venting phreatomagmatic

breccias. These exploit pre-existing structures and occur adjacent to diatreme breccia com-
plexes. Those at Busai, Woodlark Island (Corbett et al., 1994a), display a gradation from coars-er
angular fragments at depth to "flinties" at higher levels, characterised by fine rock flour and
chalcedonic silica. These breccias prepare structures for later mineralization.

Endogenous domes are indicative of the flow dome association for diatreme breccias and typi-
cally form about diatreme margins (Fig. 3.18). Many are dismembered and vary to dykes and
intrusive fragment-dominated breccias. The felsic, typically dacitic, compositions reflect the
nature of the source intrusive.

Alteration derived from the hot gasses associated with the eruption of diatreme breccias is
most commonly characterised as clay-pyrite alteration of the rock flour breccia matrix. Clays
vary from higher temperature sericitic at depth, through illite and smectite at highest levels,
with local kaolinite in acid conditions, typically at surficial levels.

Mineralization follows the pre-mineral phreatomagmatic diatreme event which taps the top of
the magma chamber at depth and fractures the overlying country rocks. Pregnant fluids which
evolve from the source magma, as it cools and degasses following emplacement, rise into the
fractured overlying country rocks, and locally the lower portion of the diatreme. The setting of
diatreme-associated mineralization is in part governed by the level of erosion and structural
environment. Diatremes formed at higher crustal levels are dominated by incompetent low
temperature clay alteration which does not fracture well, and so these rocks tend to be poorly
mineralized (e.g., Gold Ridge, Solomon Islands). Because of proximal relationship to the
magmatic source and the ability of competent sericitic clays to fracture, only deeper levels of
diatremes tend to host fracture/disseminated gold mineralization (e.g., Montana Tunnels, USA,
Sillitoe et al., 1985).

Mineralization more typically occurs within competent fractured country rocks about the di-
atreme margins, commonly at the intersection with throughgoing structures (e.g., Tolukuma, Fig.
7.50; Kerimenge, Figs. 7.28, 7.30; Lepanto; Fig. 6.24). The G.'W. breccia pipes rim the Balatoc
diatreme, Acupan occur at the intersection of throughgoing vein systems and demon-strate an
increased fluid flow along the diatreme margin (Damasco and Guzman, 1977; Sawkins et al.,
1979).

Mineralization occurs at the diatreme margin as:

* banded fissure veins at Tolukuma, PNG (section 7.iv.d, Corbett et al., 1994a); Wau,
PNG (Sillitoe et al., 1984); and Edie Creek, PNG (Section 7.iii.j; Lowenstein 1982);
* breccia infill in the G.W. breccias, Acupan, Philippines (Damasco and Guzman,
1977; Sawkins et al., 1979); and Lepanto, Philippines (Section 6.iv.b, Garcia 1991);
* stockwork veins at Kerimenge, PNG (Section 7.iii.j, Hutton et al. 1990), or
* disseminated within the diatreme at Montana Tunnels, USA (Sillitoe et al., 1985) and
Gold Ridge, Solomon Islands (Section 7.iii.j; Fig. 7 in Sillitoe 1989).

48
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization* Corbett G J & Leach T M, 8/96 Edn.

3. Phreatic Breccias

Phreatic (meaning water converted to steam) breccia systems occur as vapour-driven explo-
sions at elevated crustal levels (Fig. 3.19), and are broadly equivalent to the hydrothermal ex-
plosion breccia of Baker et al., (1986) and eruption breccia of Nelson and Giles (1985), and
Hedenquist and Henley (1985). Most workers (Phillips, 1973; Sillitoe, 1985; Nelson and Giles,
1985; Baker et al., 1986; Hedenquist and Henley, 1985) provide a mechanism for brecciation
based upon violent release following a build up of hydrostatic pressure, typically within geo-
thermal terrains. Their models mostly focus upon the development of impermeable barriers
through silica deposition by rapid pressure release during previous eruptions. Reactivation of
existing structures which may have focused hydrothermal fluids, by earthquakes and magma
intrusion (Sillitoe, 1985: Hedenquist and Henley, 1985), may initiate eruption by fracturing the
silicified cap to the overpressurised hydrothermal fluids.

Examples of phreatic or eruption breccias in the southwest Pacific are confined to little eroded
settings such as the active geothermal districts of the Taupo Volcanic Zone, New Zealand, and
associations with mineralization are described more fully in section 8.vii. Anomalous gold and
other metals occur at eruption breccia vents at Champagne Pool, New Zealand (Hedenquist and
Henley, 1985), and the Beppu district and Osorezan, Japan (section 8.vii; Aoki, 1989).
Unmineralized eruption breccias at the Ladolam deposit Lihir Island probably relate to the re-
cent geothermal activity (Fig. 7.5). Fossil geothermal terrains host mineralization associated
with eruption breccias at the McLaughlin mine, and other hot spring deposits of western USA
(Section 8.vii; Lehrman, 1986; Nelson and Giles, 1985), Puhipuhi, New Zealand (White,
1986), the Yamada veins, Hishikari, Japan (Izawa et al., 1993) and at Toka Tindung, North Su-
lawesi (Wade, 1996).

Eruption craters or eruption breccia vents vary from a few to typically tens of metres wide by
up to several hundred metres deep and commonly lie on regional structures (Fig. 3.19). As
these represent surficial to shallow level features, preservation is commonly restricted to
young, poorly eroded terrains. Many craters act as outflows for typically neutral chloride flu-
ids, which mix with cool surficial waters to promote the deposition of siliceous sinters (Fig.
8.1). The less common bicarbonate fluids form travertine deposits. At Champagne Pool in New
Zealand, eruption breccia vents are localised on fractures which display angular relatio n-ships
to regional structures and are interpreted to have been dilated by regional strike-slip rota-tion
(Fig. 8.5). Many eruption breccia systems are emplaced in environments of advanced ar-gillic
clay alteration (Section 8.v).

Eruption breccias vary from ejecta projected over considerable distances, to insitu brecciation
and introduction of a matrix of hydrothermal fluid. In the former case, breccias tend to be
massive, poorly sorted (Nelson and Giles, 1985) and supported matrix of milled rock material
(Hedenquist and*Henley, 1985). Angularity is dependent upon the degree of milling during
transport, and fragment styles are dominated by the rocks through which the eruption has
passed. Although lacking the juvenile porphyry fragments which characterise diatreme breccia
bodies, eruption breccias may host exotic ejecta. At Osorezan, Japan, where precipitates from
recent drilling contain antimony, arsenic and mercury, eruption breccias contain ejected frag-
ments of bladed stibnite and auriferous banded quartz (Aoki, 1989; Section 8.vii). Many more
locally-derived breccias exhibit monolithic and angular fragments. Eruption breccias which
form fluid outflows are characterised by a predominance of sinter fragments (e.g., Phuipuhi,
Toka Tindung). Insitu brecciation results in the formation of breccias classified herein as hy-
drothermal injection breccias (section 3.viii.c.l), and similar to the vent breccias of Nelson and
Giles (1985, Fig. 1, A and B), or the hydraulic fracturing of Hedenquist and Henley (1985, Fig.
11). Here, breccia matrix are infilled with rock flour and sulphides such as

49
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

pyrite.

Alteration in eruption breccias characteristically occurs as a flooding of silica and varying de-
grees of fine grained commonly massive pyrite. This silicifiction, which is most common in flu-
id outflows, contracts with the clay and disseminated crystalline pyrite alteration in diatreme
breccias.

Mineralization is described from matrix in eruption breccias in the Nansatsu gold deposits of
Japan and hot spring-style gold deposits of Nevada by Nelson and Giles (1985). Sinter depos-its
which form as surficial outflows of fluid from eruption breccia vents typically display anom-
alous mercury, gold, silver, antimony and arsenic contents. Mineral deposition is promoted by
sudden quenching as hydrothermal fluids mix with surficial waters adjacent to the breccia vent,
whereas sinter deposits further from the vent tend to be barren. The precipitate adjacent to
Champagne Pool, New Zealand has yielded anomalous gold to 80 ppm and silver to 175 ppm
(Weissberg, 1969) and explosion breccia-related sinter deposits from the Kinruyu Hot Spring,
Beppu, Japan yielded; 0.47 ppm Au, <1 ppm Ag, 12.2 percent As, 1290 ppm Sb and 9990 ppm
W (Austpac Gold unpubl. data). The McLaughlin gold mine, USA (Tosdal et al., 1993) and
Puhipuhi gold prospect, New Zealand (Brown, 1989; White, 1986) each have recorded histo-ries
of mercury production from sinter deposits adjacent to explosion breccia vents and have since
undergone exploration for underlying gold mineralization, leading to production in the ease of
McLaughlin (Section 8.vii).

Much of the gold mineralization mined at McLaughlin lies mostly within veins below the ex-
plosion breccia (Sherlock, 1993; Tosdal et al., 1993). Similarly, the setting of the Yamada veins,
Hishikari, Japan, immediately underlying the explosion breccia (Izawa et al., 1993), may have
promoted downward movement of groundwaters and mineral deposition by quench-ing
(Section 8.vii). Vein mineralization is interpreted to also underlie explosion breccias at
Osorezan, Japan (Aoki, 1989) and at Toka Tindung, North Sulawesi (Wade, 1996). Thus, gold
mineralization is most likely to occur below eruption breccias (Sillitoe 1985), as veins of the
stockwork, sheeted or fissure vein type, or matrix to breccias.

ix) Conclusion

Pacific rim gold-copper systems require fractured rocks to provide open space for the flow of
magmatic mineralized fluids. Mineral deposition may be promoted by the mixing of these flu-
ids with ground waters present within the same plumbing systems. Differing styles of frac-tured
host rocks may be modelled by comparisons with: active fault systems, experimental da-ta, and
field observations of many mineralized systems. Models may point towards the better
mineralized portions of hydrothermal systems and the best orientation for drill testing.

50
Exploration Workshop "SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralisation" Corbett G J & Leach T M, 8/96 Edn.

Fig. 4.4

Fig. 4.5
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

c) Sulphate minerals

The solubilities of sulphate minerals are also strongly controlled by temperature, and like car-
bonate minerals exhibit an inverse relationship with increasing temperature (Fig. 4.4). Gypsum
deposits in preference to anhydrite at low temperatures (<100-150°C). In highly saline porphyry
environments, anhydrite solubility decreases proportionally to temperature, causing the common
occurrence of (biotite) anhydrite in porphyry environments. At high activities of sulphur, and/or
low fluid pH, alunite will deposit in preference to anhydrite (Reyes, 1985).

Barite is the least soluble of the sulphate minerals and will preferentially deposit upon heating at
low temperatures of a barium-bearing solution (e.g., circulating sea water). However, unlike
other sulphate minerals, barite exhibits a reversal in solubility in relation to temperature at higher
temperatures (150-200°C in very dilute fluids), and therefore becomes increasingly more soluble
at higher temperatures (Fig. 4.4). In the saline environment of upwelling hot high sulphidation
systems, barite solubility decreases with temperature and is therefore commonly deposited in the
central, highly silicified zones.

v) Controls on Metal Deposition

a) Gold
2
Gold is transported as a bisulphide complex (Au(HS)' ) over almost all conditions within meso-
o
thermal and epithermal environments (<300-350 C), and probably as a chloride complex
(AuCl") at higher temperatures (Fig. 4.5; Seward, 1982). Gold is transported as a bisulphide
complex over pH ranges characteristic of low sulphidation systems (pH >4), and as chloride
complex under the acidic environment of high sulphidation systems (Large 1994; Fig. 4.5). This
is supported by the common occurrence of gold with copper mineralization in high sul-phidation
deposits (Section 6.i.e). AuHS0 has also been identified as one of the principal gold complexes in
high temperature acidic solutions (Bening and Seward, 1994; Arribas, 1995). Elsewhere
(Seward, 1982), it has been postulated that under low pH (<3-4) conditions gold may be
transported as either a thio-bisulphide or as a combined chloride-bisulphide complex.

In high temperature porphyry environments gold is transported as a chloride complex and is

commonly intergrown with copper phases. Its deposition is controlled by decreases in tempera-
ture, pressure, and salinities. At lower telnperatures, gold is transported as a bisulphide complex
and its solubility actually increases with increasing temperature and salinity (Henley 1985a;
Figs. 4.5, 4.6), [although Seward (1982) illustrated that gold solubility actually decreases with
cooling]. Gold mineralization in these environments can be given by the following equation:
2 +
Au + 2H2S = Au(HS)- + 1/2H2 + H
Therefore, in mesothermal to epithermal environments gold will be deposited from an
upwelling mineralized fluid upon:
1. Boiling,
2. Mixing with low pH fluids,
3. Mixing with oxygenated fluids.

1. Boiling causes exsolution of H2S and associated gold deposition. However, this is somewhat
modified by an associated increase in pH and decrease in temperature which actually increase

58
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization* Corbett G J & Leach T M, 8/96 Edn.

gold solubility_during boiling (path A in Fig. 4.6). Brown (1986) illustrated that flashing of deep
hydrothermal fluids to atmospheric pressure at Broadlands Geothermal Field, New Zea-land,
deposited percentages of gold and silver in scales in surface pipework. However, only 14 percent
of gold calculated to be in the deep fluid was deposited, due to effects of temperature loss, pH
increase, and gold completing with bisulphide remaining in fluid after boiling. The extreme
pressure drops in these experiments are unrealistic in natural environments, and there-fore are
considered to illustrate (only) that boiling can deposit some gold, under unique cond i-tions (see
Sections 7 and 8). The decrease in activity of sulphur is an important factor which promotes gold
mineralization, and can also be achieved by deposition ofjibundant sulphides (e.g., Fe-sulphides
in high sulphidation systems).
2. Mixing with low pH fluids such as descending acid sulphate or bicarbonate fluids will
deposit gold (path C in Fig. 4.6). However, this is offset by dilution and cooling effects.

3. Oxidation of a fluid by mixing with descending oxygenated groundwaters is an excellent

mechanism for the deposition of gold (path B in Fig. 4.6). Slight drops in the fO2 of ascending
mineralized fluids in the region of pyrite/hematite phase boundaries can decrease gold
solubility by orders of magnitude,(Romberger, 1988; Fig. 4.6). The oxidation of pregnant fluids
in an epithermal environment is therefore probably the most effective mechanism of gold
deposition, and is considered the main mechanism for the formation of bonanza-grade gold
deposits. This is supported by the common occurrence of hypogene hematite in these deposits
(Sections 7 and 8).

[Under low pH (<3-4) conditions, especially at high temperatures, the solubility controls on
gold are poorly understood. Gold may be transported as the thio -bisulpide complex
HAu(HS)2°, or as combined chloride-bisulphide complexes (Seward, 1982)].

b) Copper

Copper is transported as a chloride complex (either as CuCl0 or CuCl32") over temperature and
pH ranges which are encountered in most hydrothermal systems (Fig. 4.5; Large, 1994). Cop-
per minerajization is therefore controlled by decreases in temperature, salinity and pjigiiuxe7
Ifhas been postulated (Henley and Brown, 1985) that copper may also be transported
b i l J d i at low temperatures and salinities. This is supported by the common occurrence of
low grade, but significant, copper which accompanies gold mineralization in many
epithermal quartz vein deposits (see Section 8).

c) Lead and Zinc

These metals are transported as chloride complexes under most hydrothermal conditions, and
their deposition is caused by decreases in temperatures (Fig. 4.5; path A in Fig. 4.7), salinities
(path B in Fig. 4.7) and pressure (Henley, 1985a, 1985b).

d) Silver

Silver is transported mainly as a bisulphide in dilute conditions, and as a chloride complex in

hotter more saline conditions (Fig. 4.7). Therefore, silver mineralization mimics base metal
phases at deeper levels and gold at shallower, epithermal levels. Most high sulphidation sys-tems
are remarkably low in silver implying either that hot, very acidic fluids, are Ag-depleted, or as is
more likely the case, Ag-complexes are much more soluble than gold under low pH conditions
(Fig. 4.7). Silver mineralization is however significant in the Wetar high sulphida-tion deposit,
Indonesia (Sewell and Wheatley, 1994), which possibly reflects an

59
Exploration Workshop 'SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralisation" Corbett G J & Leach T M. 8/96 Edn.

Fig. 4.6
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

increase in Ag-solubility in low pH environments associated with circulating saline sea water,

e) Gold Fineness

The silver content of gold is commonly given as the fineness (=Au/Au+Ag x 1000). Previous
workers (Morrison et al., 1991) illustrated that Archean, Slate Belt and Plutonic systems dis-play
high and consistent fineness, and porphyry, volcanogenic and epithermal classes exhibited wide
fineness variations within each deposit and in the overall types of systems. These apparent wide
variations may be explained by:
* widely differing environments in a single class (e.g., Mt. Kasi high sulphidation
system with Hishikari adularia-sericite epithermal system; Wau shallow level
carbonate-base metal system with Guinaoang porphyry copper-gold system,
* bullion data (e.g., Waihi) which incorporates silver from phases other than gold, with
microprobe data on gold grains.

Figure 4.8 illustrates histograms of the fineness of gold on selected southwest Pacific gold ±
copper prospects and deposits, based on the classification used for these systems in this manu-al.
The data is derived from unpublished microprobe analyses by Ken Palmer (Victoria Univer-sity,
Wellington, NZ), except where indicated.

There is a systematic decrease in gold fineness at progressively cooler and shallower levels (or
distal from the intrusive source), from:
* porphyry copper/skarn systems (average = 920),
* through quartz sulphide (average = 850) and
* carbonate-base metal (average = 765) systems, to
* epithermal quartz silver-gold systems (average = 685).

These zonations therefor indicate that temperature is probably the most important control on
gold fineness in intrusion-related gold-copper systems. Wide ranges in gold fineness within a
single deposit is interpreted to reflect wide temperature ranges during mineralization: (e.g., at
Kidston gold is associated with both quartz-sulphide and carbonate-base metal styles of vein-ing;
at Mt Kare gold in the roscoelite-bearing quartz silver-gold systems was deposited over a very
wide temperature range (approx. 250°C to >100-150°C).

The fineness of gold in intrusion-related epithermal quartz silver-gold systems, and epither-mal
adularia-sericite systems is closer to the fineness of gold in carbonate-base metal systems than
would be expected from the generally cooler conditions of mineralization. However, many
carbonate-base metal systems form at very shallow epithermal levels (e.g., Maniape, PNG;
Karangahake, New Zealand) and therefor the gold is silver-rich. In addition, the prefer-ential
partitioning of silver into other phases (Ag-sulphosalts, sulphides and tellurides) in epi-thermal
systems could result in less silver being available to be incorporated into the native
gold/electrum (Afifi et al., 1988). The similar fineness of gold in intrusive-related quartz gold-
silver and adularia-sericite epithermal gold-silver systems reflects comparable environ-ments of
gold mineralization, and also possibly the magmatic source to the metals in both sys-tems.

In high sulphidation systems, free gold is only observed in very shallow, epithermal enviro n-
ments. However, even under these cool conditions the fineness of gold is very high (average of
935 for Mt. Kasi, Peak Hill and Zone A at Wafi River). The silver -undersaturated nature of
most fluids in southwest Pacific high sulphidation systems is difficult to reconcile and re-quires
further study.

60
Exploration Workshop 'SW Pacific Rim Au/Cu Systems: Structure Alteration & Mineralization' Corbett G J & Leach T M, 8/96 Edn.

Histograms of gold fineness in southwest Pacific gold - copper prospects and

deposits (microprobe analyses by Ken Palmer except where indicated)
Figure 4. 8
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

4 CONTROLS ON HYDROTHERMAL ALTERATION AND MINERALIZATION i)

Introduction

The large number of variables which affect the formation of alteration minerals in hy-
drothermal systems are grouped into six main factors (Browne, 1991):
Temperature
1. Fluid chemistry
2. Concentrations
3. Host rock composition
4. Duration of activity or degree of equilibrium.
5. Permeability

Although these are all more or less interdependent, temperature and fluid chemistry probably
display the greatest influence on the styles of hydrothermal alteration.

Increasing temperature favours the stability of progressively more dehydrated mineral species.
This is especially evident in clay/sheet silicate mineralogy in which the progressively higher
temperatures result in the mineral sequence: smectite, interlayered smectite -illite (with gradu-
ally decreasing smectite content), illite and white mica. Similarly, zeolites become more
dehydrated under hotter conditions, as illustrated by the sequence mordenite -------- > stilbite
— —> laumontite ------ > wairakite.

Temperature also affects the degree of ordering or crystallinity of minerals. Higher tempera-
tures favour the formation of more crystalline phases. Disordered kaolinite forms under ambi-
ent conditions, whereas more ordered kaolinite occurs under elevated hydrothermal tempera-
tures, and well crystalline dickite develops under still hotter conditions.

It can be seen from activity diagrams that fluid composition also has a strong influence on the
alteration mineralogy, with temperature having a marked affect on the position of phase bound-
Nl H+
aries. More important than absolute concentrations are the ratios of constituents e.g., a 7a ,
K H
a 7a \ For example, in active hydrothermal systems, highly saline brines of the Salton Sea geo-
thermal field (approximately 250,000 ppm total dissolved solids) produce the same alteration
assemblages at most temperature ranges as the very dilute fluids (around 3,000 ppm total dis-
solved solids) of New Zealand, and some Icelandic geothermal fields.

Absolute concentrations in hydrothermal fluids have some affect on the type of alteration
mineralogy, since this affects the degree of saturation of the fluid with respect to certain min-
erals. For example, sulphur, sulphides, and/or sulphates are associated with solfataras, and le-
pidolite is encountered in the Salton Sea geothermal fields where fluids have high concentra-
tions of lithium.

The host rock composition to some extent controls the type of alteration mineralogy. Skarn
mineralogy forms in calcareous host rocks. The secondary K-feldspar phase adularia is prefer-
entially encountered where host and/or source rocks are potassium-rich (e.g., rhyolite or sho-
shonite). Paragonite (Na-mica) under certain conditions forms as an alteration product of al-bite,
whereas muscovite forms from altered potassic feldspars. The kinetics or rates of altera-
tion/deposition more commonly affect the crystallinity of the minerals rather than the species
formed. Amorphous silica can form at moderately high temperatures where silica-saturated flu-
ids are quenched (e.g., geothermal surface pipework), whereas coarse

51
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

crystalline quartz forms at the same temperature but under static conditions which permit slow
crystal growth.

The duration of the hydrothermal system, or the period during which permeability has re-mained
open, determines whether equilibrium has been established between the circulating fluid and
host rocks.

Permeability has the obvious effect of bringing the host rock into contact with circulating fluid.
Phyllic and argillic alteration are commonly encountered immediately adjacent to major struc-
tures or vein systems where the fluid is at less than neutral pH, due to dissolved gases, where-as
propylitic alteration is usually encountered in low-permeability host rocks some distance from
the main fluid channelways.

ii) Temperature and pH Controls on Alteration Mineralogy

Temperature and fluid pH are the most important of many factors which influence the miner-
alogy of hydrothermal systems. Pressure is directly related to temperature, whereas the gas
pressure and the ratios of elemental concentrations can be expressed in terms of pH. The other
variables (with the local exception of perhaps host rock composition and absolute composi-
tions) have only minor effect on alteration mineralogy.

The most common hydrothermal minerals encountered in Pacific rim active geothermal and
hydrothermal ore systems, in a variety of temperature and fluid pH conditions are illustrated in
Figure 4.1. This interpretation is derived from a compilation of data from geothermal systems in
the Philippines, Japan, USA, Iceland, and New Zealand, in combination with thermodynam-ic
and laboratory experimental work on various mineral phases. Although fluid element co n-
centrations and ratios, and pressures (gas, hydrostatic and lithologic) have been kept constant, in
many cases these factors can substantially effect mineral stability ranges. Discussion of var-
iations in these factors, which would add further axes to this diagram, is beyond the scope of
this section. Absolute temperature and pH values have not been included in the axes of Figure
4.1 because of the effects of these other variables on the position of the boundaries between
mineral phases. However, the following discussions include the approximate temperature and
pH ranges for most of these mineral phases. Different mineral groups categorized by increas-ing
pH of formation on Figure 4.1 are:

a) Silica group minerals

Silica phases are the only significant stable alteration minerals at very low pH (generally be-low
pH 2, Hedenquist et al., 1988), commonly associated with small amounts of titanium-iron
phases such as rutile. Under these extremely acid conditions, opaline silica, cristobalite, and
tridymite are encountered within surficial environments above the hydrothermal water level,
typically at temperatures of <100°C (Leach et al., 1986). Quartz is the main silica phase at high
temperatures. In Figure 4.1, quartz or silica (cristobalite, tridymite or amorphous silica) has been
included in all mineral assemblages because hydrothermal fluids (in active geother-mal systems)
are most commonly saturated with respect to SiO2. Exceptions occur in environ-ments where
hydrothermal systems are hosted in silica-poor rocks (e.g., basalts, phonolites).

Under higher fluid pH conditions, amorphous silica is encountered at cool conditions. Quartz
is almost ubiquitous at higher temperatures, whereas chalcedony locally occurs at intermediate
temperatures (generally in the range of 100-200°C), especially under conditions of rapid

52
Mineral Abbreviations:
Ab - albite; Act - actinolite; Ad - adularia; Al - alunite; And - andalusite; Cb - carbonate (Ca, Mg, Mn, Fe)
Ch - chlorite; Chab - chabazite; Chd - chalcedony; Ch-Sm - chlorite-smectite; Cor - corundum; Cpx -
clinopyroxene; Cr - cristobalite; Ct - calcite; Do - dolomite; Dik - dickite; Dp - diaspore; Ep - epidote;
Fsp - feldspar; Ga - garnet; Hal - halloysite; Heu - heulandite; I - illite; I-Sm - illite-smectite; K - kaolinite;
Lau - laumonite; Mt - magnetite; Mor - mordenite; Nat - natrolite; Op - opaline silica; Pyr - pyrophyllite;
Q - quartz; Ser - sericite; Sid - siderite; Sm - smectite; Stb - stilbite; Tr - tremolite; Tri - tridymite;
Ves - vesuvianite; Wai - wairakite; Wo - wollastonite; Zeo - zeolite

Figure 4.1 Common Alteration Mineralogy in Hydrothermal Systems

deposition. The type of silica phase is also affected by the kinetics of deposition. For example,
amorphous silica may be formed at temperatures up to 200°C in rapidly quenched enviro n-
ments (e.g., scales in geothermal surface pipeworks; Brown, 1986).

b) Alunite group minerals

At fluid pH slightly higher than 2, alunite is formed together with the silica phases over a
wide temperature range, in association with andalusite at high temperatures (typically >
300-350°C), and corundum occurs at still higher temperatures (Hemley et al., 1980).

Four environments of alunite formation have been identified by Rye et al., (1992), using sul-
phur and oxygen isotope data. Different styles of alunite can also be identified using simple
crystal textures in conjunction with associated alteration minerals and the geological setting:

1. Steam-heated alunite develops under surficial environments by the oxidation of acid

sulphate fluids from H2S gas evolved from a boiling hydrothermal system at depth. This
steam-heated alunite may be encountered down to 1-1.5 km depths, in settings where acid
sulphate fluids descend into waning hydrothermal systems. It occurs as very fine-grained,
poorly crystalline acicular crystals.

2. Supergene alunite develops from the production of sulphuric acid by weathering of massive
sulphide deposits, also occurs as poorly crystalline acicular crystals, and may be distinguished
from steam-heated alunite by its geological setting and common association with iron oxide
weathering products.

3. Magmatic alunite is deposited from volatiles ascending directly from an intrusive source
and commonly occurs in veins and breccia zones, as radiating prismatic crystals. Alunite
formed in close proximity to the parent porphyry can also occur as large irregular crystals
poikilitically enclosing quartz and other phases, or as euhedral rhombic crystals.

4. Liquid alunite is derived from magmatic-dominated liquids and forms well crystallised,
commonly coarse-grained tabular to lath-like crystals infilling fractures, breccia zones, and
leached vughs pseudomorphing phenocrysts or lithic clasts.

c) Kaolin group minerals

The kaolin group of minerals (Fig. 4.1) are derived from higher pH fluids (approximately pH 4),
and co-exist with the alunite-andalusite-corundum group of minerals under a transitional fluid
pH range (3-4). Halloysite occurs mainly as a supergene weathering product, although there is
some evidence (Harvey and Browne, 1991) for hydrothermal-derived halloysite. Zona-tions of
hydrothermal kaolin group minerals with increasing depth and temperature have been identified
in Philippine geothermal systems by Reyes (1990b) and Leach et al. (1986). Kaolin-ite is
formed under shallow, low temperature conditions (<150-200°C) and pyrophyllite under deep,
higher temperature conditions (<200-250°C), whereas dickite occurs transitional be-tween these
two levels and temperature ranges.

Diaspore is locally encountered in intensely silicified zones with alunite and/or kaolinite group
phases. Hemley et al. (1980) indicated that diaspore is formed under conditions of silica un-
dersaturation. Since diaspore occurs in zones of intense silicification it is assumed that its
formation results from the rapid precipitation of quartz, producing a silica-undersaturated so-
lution.

53
 Exploration Workshop 'Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

d) lllite group minerals

At conditions of progressively higher fluid pH (4-6 range) the illitic group of minerals (Fig. 4.1)
becomes the dominant phase, co-existing with the kaolin group minerals at transitional fluid pH
(4-5). Depth/temperature relationships of the illite group minerals are well document-ed from
both sedimentary basins and active geothermal systems (Steiner, 1977; Browne, 1991; Harvey
and Browne, 1991). Smectite occurs at low temperatures (<100-150°C), interlayered illite-
smectite at around 100-200°C, illite at approximately 200-250°C, well-crystalline fine-grained
mica (sericite) at >200-250°C, and coarse crystalline white mica (typically phengite) at >250-
300°C.

The smectite content within the interlayered illite-smectite clays decreases progressively with
increasing temperature over the 100-200°C range (Harvey and Browne, 1991). The crystallinity
of illite and sericite increases with increasing temperature, and can be monitored by XRD anal-
yses on the peak width, at half the peak height, of the {001} reflection, (i.e., the Kubler I ndex).
Sericite is basically fine-grained muscovite, and both grain size and crystallinity increase at
higher temperatures. The changes in sericite/muscovite crystallinity can also be monitored by
XRD analyses, with progressive changes from a disordered 1M mica to a well crystallized 2M
muscovite with increasing temperature. Although muscovite is the common illite/mica phase
present, the sodic phase paragonite is encountered in some systems where the host rock has a high
Na:K ratio (e.g., albite as the plagioclase phase). The vanadium mica phase roscoelite, and the
chromium phase fuchsite, are deposited from fluids which had source, or migrated through, basic
volcanic/intrusive rocks.

e) Chlorite group minerals

Under (slightly acid to) near neutral pH conditions chlorite-carbonate (Fig. 4.1) phases be-come
dominant, coexisting with illite group minerals in transitional environments (pH 5-6; Leach and
Muchemi, 1987). Interlayered chlorite-smectite occurs at low temperatures, grading to chlorite at
higher temperatures. This transition is encountered at different temperatures in active geothermal
systems within different geological settings. Chlorite occurs at significantly lower temperatures
in rift environments (e.g., Iceland, Kristmannsdotter, 1984) than in volcanic island terrains (e.g.,
Philippines, Reyes, 1990a), possibly in response to the effects of either flu-id or host rock
chemistry. (Chloritic clays co-exist with illitic clays under transitional fluid pH values).

f) Calcsilicate group minerals

The calcsilicate group of minerals (Fig. 4.1) form under neutral to alkaline pH conditions. Ze-
olites-chlorite-carbonate occur at lower temperatures, and epidote, followed by secondary am-
phiboles (mainly actinolite) develop at progressively higher temperatures. Zeolite minerals are
particularly temperature sensitive. Hydrous zeolites (natrolite, chabazite, mordenite, stilbite,
heulandite) form under cool conditions (<150-200°C), while less hydrated zeolites such as
laumontite (150-200°C), and wairakite (200-300°C) occur at progressively deeper and hotter
levels in the hydrothermal system (Steiner, 1977; Leach et al., 1983). In some systems prehnite
(250-300°C; Elders et al., 1982) or pumpellyite are encountered in association with, or in places
instead of, epidote.

Epidote occurs as poorly crystalline incipient grains at temperatures of around 180-220°C, and
as better crystalline phases at higher temperatures (>220-250°C). Secondary amphiboles
(mainly actinolite) appear to be stable in active hydrothermal systems at temperatures >280-
300°C (Leach et al., 1983). Biotite is commonly ubiquitous within or immediately adjacent to

54
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

porphyry intrusions, and in active systems occurs at temperatures of >300-325°C (Elders et al.,
1982; Leach et al., 1983). Active porphyry environments are characterised by clinopyroxene
(>300°C) and garnet (>325-350°C) assemblages (Elders et al., 1982). However, hydrated gar-
nets are locally encountered at significantly lower temperatures (250-300°C) in the Tongonan
geothermal field (Leach et al., 1983). The zonations in skarn mineralogy with temperature are in
many ways comparable to those in porphyry copper environments, and are discussed in more
detail in Section 5.H.

g) Other mineral phases

Carbonate minerals are encountered over a wide range of pH (> 4) and temperature, and are
associated with kaolin, illite, chlorite and calc-silicate phases. A zonation in carbonate species
with increasing fluid pH is encountered in many hydrothermal systems (Leach and Corbett,
1993, 1994, 1995) as: Fe-Mn carbonate (siderite-rhodochrosite) coexist with kaolin and illitic
clays, mixed Ca-Mn-Mg-Fe carbonates (rhodochrosite-ankerite-kutnahorite-dolomite) occur
with illitic and chloritic clays, and Ca-Mg carbonates (dolomite-magnesian calcite-calcite) co-
exist with chlorite-calcsilicate mineralogy. This zonation is interpreted to reflect the decreas-ing
mobility of Fe, Mn and Mg at progressively increasing fluid pH (Leach et al., 1986). Car-bonate
minerals typically extend throughout all levels in hydrothermal systems, from surficial
environments to porphyry-related skarn environments.

Feldspar minerals are associated with both chlorite and calcsilicate mineral phases. Secondary
feldspars are generally stable under near neutral to alkaline pH conditions. Albite occurs where
fluids have a high aNl7aK+ ratio and potassium feldspar a low a Nl7aK+ ratio (Browne, 1978).
Adular-ia occurs as a low temperature secondary potassium feldspar species, whereas orthoclase
is en-countered at high temperatures within the porphyry environment. Browne (1978)
demonstrated that adularia preferentially occurs within high fluid flow permeable conditions,
and albite under low permeability conditions.

Sulphate minerals are encountered over most temperature and pH ranges in hydrothermal sys-
tems. Whereas alunite forms under low pH (<3-4) conditions, anhydrite forms at a higher pH
(Reyes, 1985) and temperatures greater than 100-150°C, and gypsum develops in cooler envi-
ronments (Harvey et al., 1983). Although jarosite commonly forms as a weathering product of
sulphides, it also occurs at shallow levels in acid environments in some Philippine active geo-
thermal systems (Leach et al., 1986).

Various hydrothermal mineral phases contain halogen elements (e.g., boron-tourmaline; fluo-
rine, chlorine and phosphorous-apatite), which are indicative of fluids which contain a signifi-
cant component of magmatic volatiles. These phases are commonly associated with serici-
te/mica formed at high temperature and moderately low pH conditions.

iii) Alteration Zones Associated with Ore Systems

Alteration assemblages within Pacific rim gold-copper hydrothermal systems have historically
been divided into seven main alteration types: advanced argillic, argillic, phyllic, propylitic,
sub-propylitic, potassic and skarn. Mineral assemblages are commonly assigned to each of
these broad alteration zones with fairly arbitrary boundaries. Although it is considered best to
use the alteration mineral assemblages themselves in defining the style of alteration, the clas-
sification of these broad alteration types can be beneficial in describing overall characteristics
of zoned alteration systems.

55
The mineral assemblages generally assigned to these seven alteration zones have been outlined
in Figure 4.1 as:

Advanced argillic alteration consists of the low pH silica, alunite and transitional
alunite-kaolin group of minerals.

Argillic alteration assemblages include members of the kaolin (halloysite, kaolinite, and dickite)
and illite (smectite, interlayered illite-smectite, illite) group of minerals and associated
transitional mineral assemblages, which formed at intermediate pH and low temperature. Tran-
sitional low temperature chlorite—illite group mineral assemblages may also be termed argillic.

alteration includes members of the kaolin (pyrophyllite-andalusite) and illite (sericite-

white mica) mineral groups, and associated transitional assemblages, which along with sericite-
mica-chlorite assemblages formed in higher temperature conditions.

Subpropylitic alteration zones include low temperature chlorite-zeolite, and with propylitic al-
teration are generally assigned to higher temperature chlorite-epidote-actinolite alteration as-
semblages. In some cases actinolite-bearing assemblages have been termed "inner propylitic".

Potassic alteration zones are those which contain biotite-K-feldspar-actinolite +

clinopyroxene mineral assemblages.

Skam mineralogy consists of zoned calc-silicate mineralogy such as Ca-garnet,

clinopyroxene, and tremolite.

iv) Controls on the Deposition of Gangue Mineral Phases

The main non-ore or gangue mineral phases which are deposited directly from solution in ore-
forming systems are silica minerals (predominantly quartz) and carbonate minerals, with local
abundances of sulphate mineral species. The following sections outline the main factors
which control the deposition (and dissolution) of these mineral phases in hydrothermal sys-
tems.

a) Silica minerals

Temperature is the major control on the deposition of silica mineral species, with subordinate
influence of pressure, salinities, pH, kinetics of deposition, and complexing agents (Fournier,
1985a). In epithermal to mesothermal environments (i.e., <300-350°C) quartz will deposit up-
on cooling, with a solubility maximum occurring around 350°C (Fig. 4.2). Therefore, rapid
quenching in the upper regions of an upflow hydrothermal fluid will produce a silica cap to that
system. Between 300-350°C, decreases in pressure and salinity have a moderate effect on
quartz deposition, whereas below 300°C, the effects are minor except under drastically chang-
ing conditions.

At shallow levels in an epithermal system (<100-150°C) various silica mineral species may be
deposited, including amorphous silica, cristobalite, tridymite, chalcedony and quartz. Although
quartz is the least soluble of all the mineral phases, the rates of change in temperature control
the mineral species formed. For example, amorphous silica is deposited as silica sinters under
rapidly cooling conditions, where boiling hydrothermal fluids at <200°C rapidly cool to less

56
Fig. 4.2

Fig. 4.3
 Exploration Workshop 'Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization' Corbet! G J & Leach T
M, 8/96 Edn.

than 100°C in surficial environments. Cristobalite and chalcedony form under progres-
sively slower cooling environments. Silica mineral species recrystallise to quartz with
time and/or increasing temperature. Fine-grained silica, especially thin colloform
banding, forms under rapid quenching conditions, whereas well-formed drusy (crys-
talline) quartz forms under slow cooling conditions within open spaces.

Under acidic conditions the reaction of low pH fluids with (feldspars and mafic phas-
es) wall rock results in the breakdown of many mineral phases and causes excessive
silica supersaturation. Therefore, silicification always accompanies acid leaching. In
addition, sulphate-silica complexes in low pH fluids facilitate silica supersaturation
(Fournier, 1985a). Deposition of sulphate minerals such as alunite or barite, may
therefore result in extensive silica deposition.

At high temperatures (>300-400°C), both pressure and fluid salinity, as well as tem-
perature have major controls on silica deposition (Fig. 4.2), such that quartz solubili-ty
increasing substantially with increasing pressure and temperatures. Therefore, rapid
pressure drops possibly associated with a change from lithostatic to hydrostatic pres-
sures in a porphyry environment (see section 5.i.c.ii) result in excessive silica satura-
tion, and the associated development of quartz stockwork veining (Fig. 4.2). Addi-
tional quartz veining results from the mixing and subsequent dilution of this saline fluid
with circulating dilute meteoric-dominated waters.

b) Carbonate minerals

The solubility of carbonate minerals in an aqueous solution (at pH 4-8) is best given
by the equation:

The greatest control to carbonate deposition occurs with increasing temperature,

whereas dilution and pressure drop display only secondary effects on carbonate depo-
sition (Fig. 4.3; Fournier, 1985b). Therefore, carbonate deposition is promoted at
shallow levels by the heating of descending bicarbonate fluids. If a fluid, which con-
tains significant concentrations of dissolved CO2 (PCO2), boils rapidly, then release of
CO2 promotes carbonate deposition (Fig. 4.3), typically as bladed carbonate (Sim-mons
and Christenson, 1994). This release of CO2 is accompanied by an increase in fluid pH
which will inhibit quartz deposition. Bladed carbonate in many cases is sub-sequently
replaced by quartz in response to dissolution of carbonate and deposition of silica upon
late stage decrease in fluid temperatures. Thus many epithermal sys-tems exhibit quartz
pseudomorphing platy carbonate textures.

At pH <4 dissolved CO2 gas is the only carbonate aqueous species present in equilib-
rium with gaseous CO2 and so carbonates are dissolved below this pH range. With
progressively increasing pH, HCO3' dominates, and at pH >8 the CO3'2 dominates (El-
lis and Mahon, 1977).

The controls to carbonate solubility under saline conditions are not well docu-
mented. However, comparisons with sulphate solubility, and the observed deposi-
tion of carbonate minerals from magmatic brines in porphyry environments (see
Section 5), suggest that carbonate solubility may drop with decreasing tempera-
ture under saline conditions.

57
Exploration Workshop "Southwest Pacific rim gold -copper systemsiStructure. Alteration and Mineralization: Corbett GJ & Leach TM. 3/96
5 GOLD-COPPER SYSTEMS IN PORPHYRY ENVIRONMENTS

I: PORPHYRY COPPER-GOLD SYSTEMS

a) Structural setting
The major portion of the gold and most of the copper resources of the southwest Pacific
rim occur within porphyry copper -gold deposits formed in subduction related I-type vol-
canoplutonic arcs (Sillitoe, 1992). A variety of subduction related arc settings for porphyry
intrusion at plate margins are distinguished by Sillitoe (1992) and described in detail in
Section 3.ii. Although magmatic arcs which host porphyry intrusions are typical-ly
described as having formed during orthogonal or oblique convergence and associated
subduction, most arcs demonstrate changes in style through time.
The shapes of vein systems formed peripheral to many porphyry intrusions are
indicative of emplacement during a relaxation of orthogonal compression resulting in
localised ex-tension (e.g., Batu Hijau, Indonesia, Meldrum et al., 1994; and others,
Corbett, un-published, data). Similarly, porphyry emplacement is inferred to have been
promoted by changes from orthogonal to oblique compression. The Chuquicamata
porphyry copper in northern Chile is localised by a splay which is indicative of a
component of strike slip de-formation on the Falla Oeste (West Fault) (Boric et al.,
1990). The structure of many of the Ordovician porphyry-related gold-copper
occurrences in New South Wales suggests that these formed during an episode of
sinistral oblique compression (transgression) (Corbett, unpublished, data). Thus,
changes in the nature of subduction have promoted porphyry emplacement.
Major structures localise porphyry intrusions within magmatic arcs (Section 3.iii; Figs.
1.2, 3.2). Transfer structures account for variations in the dip of subducting plates and the
rates of subduction and localise melts derived from deep crustal settings and focus
overprinting intrusions. A series of transfer structures which transect the Island of New
Guinea localise porphyry intrusions at Grasberg, Porgera and Wafi (Fig. 3.3; Corbett,
1994). Arc-parallel or accretionary structures localise porphyry intrusions at splays (e.g.,
Far South East and others in the Philippines, Sillitoe and Gappe, 1984; Chuquicamata,
Chile, Boric et al., 1990; Frieda River, Papua New Guinea, Corbett, 1994) or at inter-
sections with transfer structures (La Escondida, Chile; Corbett, unpublished, data.).
Tectonic environments characterised by oblique compression and settings within splays
are ideal loci for the formation of sheeted quartz veins in porphyry-environments (Section
3.viii). Although much of the stockwork veins in porphyry systems occurs as random vein
arrays, sheeted veins formed in dilational settings provide a mechanism for the transport of
pregnant fluids from large bodies of magma into settings of ore deposition (e.g.,
Goonumbla, Eastern Australia). Porphyry systems emplaced in environments of
orthogonal compression tend to display radial fracturing indicative of the exploitation of
conjugate and other fractures (e.g., Grasberg; Kavalieris, 1994).

b) Early models of alteration and minerali zation zonation

Extensive exploration for porphyry copper deposits took place during the 1960-80 period
prior to the gold boom of the 1980s. The application of alteration zonation models devel-
oped during that time contributed towards the discovery of the Kalamazoo (Lowell,
1991a) and La Escondida (Lowell, 1991b) porphyry copper deposits.
The temporal and spatial zonations in alteration and mineralisation associated with
porphyry copper deposits is commonly attributed to a progressive change from a
hydrothermal system dominated by magmatic fluids, to one which is dominated by

61
Exploration Workshop "Southwest Pacific rim gold-copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

meteoric waters (e.g. Gustafson and Hunt,. 1975; Beane and Titley, 1981; Reynolds and
Beane, 1985). The interaction of these two cogenetic yet chemically different fluids is be-
lieved to be the cause of mineralization in response to decreasing temperature and salinity
as well as variations in pH and oxygen and sulphur fugacity (Barnes, 1979; Hemely and
Hunt, 1992). Copper mineralization in porphyry copper deposits in southwest USA has
been illustrated to have taken place in response to this mixing at temperatures of <350°C
(Nash, 1976; Beane and Titley, 1981; Reynolds and Beane, 1985).

While many of the porphyry copper deposits that occur in Chile and the western USA are
primarily copper deposits, most southwest Pacific examples such as Dizon in the Philip-
pines, Grasberg in Indonesia and Ok Tedi and Panguna in PNG, are gold rich (Titley,
1978) and so are described as porphyry copper-gold deposits (Sillitoe, 1993). Although
many porphyry copper deposits in the southwest Pacific do not contain recoverable mo-
lybdenum, it is still commonly present in appreciable levels (Titley, 1978).

No single model can adequately portray the alteration and mineralization processes
that have produced widely different styles of porphyry copper deposits (McMillan and
Pan-teleyev, 1988). Because of this, a number of models have been put forward to
illustrate the alteration and mineralization encountered in porphyry copper systems in
different geo-logical settings.
From their work on the San Manuel-Kalamazoo porphyry copper deposit in the south-
western USA, Lowell and Guilbert (1975; Fig 5.1) suggest that this porphyry copper
system exhibits zoned hydrothermal alteration assemblages which grade from centre to
pe-riphery as:
* quartz core - quartz, sericite, chlorite, K-feldspar,
* potassic zone - quartz, K-feldspar, biotite, + sericite, + anhydrite,
* phyllic zone - quartz, sericite, pyrite,
* propylitic zone - chlorite, epidote, carbonate, adularia, albite.
Sulphides are also zoned within the shell-like alteration zones and grade from a centre as:
* ore zone - pyrite, chalcopyrite, magnetite on the periphery of the
potassic alteration in contact with the phyllic alteration,
* low grade core - central lower grade equivalent of the ore shell,
* pyrite zone - pyrite » chalcopyrite which rims the ore shell within the
phyllic alteration, to a peripheral
* low pyrite shell.
Sillitoe and Gappe (1984) developed a model based on the study of 48 variably eroded
Philippine porphyry systems which are hosted in calc-alkali intrusions and volcanic
rocks (Fig 5.2). While the setting of copper-gold mineralization is much the same as in
the Lowell and Guilbert Model, the Sillitoe and Gappe Model includes features unique
to the southwest Pacific environment. These include :
* caps of advanced argillic alteration,
* diatreme breccia associations,
* intramineral intrusions,
* presence of the mineralizing system an apophysis to a late-mineral intrusion,
* the SCC alteration (defined as sericite, clay, chlorite) which occurs instead
of the phyllic (sericitic) and appears to represent transition from the potassic
(K-silicate) to advanced argillic alteration zones.
Porphyry copper deposits hosted in alkalic igneous rocks in British Columbia display
aspects of alteration and mineralization which are significantly different from those
which are hosted in calc-alkalic intrusions and volcanic rocks (Lang et al., 1995).
These differ-ences are:

62
Porphyry copper model from Lowell and Guilbert 1975
Redrawn from Lowell and Guilbert 1975

Fig. 5.1

Fig. 5.2
Exploration Workshop "SouthwesI Pacific rim gold-copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

i) alteration which is dominated by albitic (sodic) and calc-silicate (garnet, di-

opside, actinolite, epidote) minerals in addition to biotite and K-feldspar ii)
the paucity of sericitic/phyllic, argillic and advanced argillic assemblages iii)
the abundance of magnetite associated with mineralization iv) the near ab-
sence of quartz and the abundance of carbonate as alteration minerals.
Like southwest Pacific porphyry systems, the alkalic porphyry systems of B.C. are
mo-lybdenum-poor and gold-rich. Lang et al.(1995) suggest that the differences in
alteration styles between the calc-alkali-and alkali-hosted porphyry systems may be
related to dif-ferences in the composition of the fluid which is derived from magmas
of different com-positions. The Dinkidi porphyry copper deposit in the Didipio region
of eastern Luzon, Philippines, is also hosted in alkali intrusions and exhibits some of
the features described above (Garrett, 1996).
Porphyry copper deposits form within dynamic hydrothermal systems in which there are
continually changing conditions creating multiple phases of overprinting events (Gus-
tafson, 1978). This is implicit in the Sillitoe and Gappe model, since the low temperature
clays and higher temperature sericite in the SCC zone indicate that several distinct stages
of alteration took place under significantly different physical and chemical conditions.

Gustafson and Hunt (1975) developed a model for the progressive evolution of the El
Salvador porphyry copper deposit in Chile (Fig 5.3). They postulated that initial em-
placement of mineralizing porphyries at approximately 2 km depth under lithostatic
pres-sures, facilitated the formation of potassic alteration and the development of
quartz stockwork veining under very hot (>400-500°C), highly saline conditions. In
order to explain the subsequent progressive changes to lower temperature conditions
during phyl-lic and argillic to advanced argillic alteration, they proposed a change
from lithostatic to hydrostatic conditions (Fig 5.4). This change was interpreted to
have been brought about by the influx of deep circulating meteoric dominated fluids.
c) Model of polyphasal overprinting events in southwest Pacific porphyry
copper systems
Studies in active porphyry systems in the Philippines (Mitchell and Leach 1991; chapter
2 this volume) indicate that hydrothermal systems form over a prolonged timespan,
which range from an initial emplacement of the intrusion, to a sequence of events which
in-volves exsolution of fluids from the cooling melt, followed by influxes of meteoric
wa-ters at progressively lower temperatures. Some of the overprinting events in the
Philip-pine active porphyry systems result from deposition and alteration from
descending fluids of different chemistries, and temperatures from shallow levels, rather
than the insitu de-velopment of these fluids (Reyes, 1990; Mitchell and Leach 1991). The
descent of these fluids is interpreted to have been brought about by the pressure draw-
down during wan-ing of the hydrothermal system (chapter 2).
The following section describes a conceptual model for the progressive development
of porphyry copper systems in the southwest Pacific region (Figures 5.5 and 5.6). A
gen-eralized paragenetic sequence of alteration and mineralization in southwest
Pacific porphyry copper systems is illustrated in Fig 5.7.
It is not possible to present a suitable static model which represents the features encoun-
tered in all the southwest Pacific porphyry copper systems. Rather, it is considered more
beneficial to describe possible processes which are involved in the evolution of a
porphyry copper system, and then to apply these process to the development of a genetic
model which can satisfy the individual characteristics of each system.

63
Fig. 5.3
 Fig 5.5 Early stages of development of southwest Pacific porphyry copper systems.

Fig 5.6 Late stages of cooling and mineralization in southwest Pacific porphyry copper systems.
Fig 5.7 " Generalised Paragenetic Sequence of Alteration Veining and Min-
eralisation in Southwest Pacific Porphyry Cu-Au Systems

Fig 5.8
Exploration Workshop "Southwest Pacific rim gold-copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

In a similar manner to Beane and Titley (1981) and Reynolds and Beane (1985),
it is proposed that there are two major stages during the evolution of a southwest
Pacific porphyry copper system. These are:
* a 'prograde' sequence of events associated with emplacement and cooling of a
melt at shallow (<2 km depth) crustal levels, as an apophysis from a larger magma
source, and the exsolution of magmatic fluids and metals from the upper levels of
that porphyry stock. These prograde events have here been divided into the initial
formation of zoned potassic-propylitic alteration (Fig. 5.2) followed by a later
event of quartz vein development and advanced argillic alteration (Fig 5.3).

* a 'retrograde' sequence which involves the subsequent cooling of the

intrusion , to the extent that the porphyry stock (and its hosts rocks) becomes
an environment of metal deposition (Figure 5.3 Stage HI). The vast bulk of the
mineralization in porphyry copper systems in the southwest Pacific region
(and for southwest USA porphyry deposits; e.g. Cathles, 1977; Beane and
Titley; 1981, Reynolds and Beane, 1985) is interpreted to have developed
during this retrograde event at temperatures of around 250-350°C.
The changes in alteration related to variations in temperatures and fluid pH during
these events is illustrated in Figure 5.8.
The following conceptual model of the evolution of porphyry copper systems in the
south-west Pacific is therefore presented as a framework to develop site specific
models for each individual prospect or deposit.
Stage I. Zoned Potassic-Propylitic Alteration
Initial emplacement of melts at shallow crustal levels and the associated cooling and crys-
tallisation is accompanied by the formation of zoned alteration assemblages formed in re-
sponse to the transfer of heat from the melt into host lithologies (McMillan and Pantele-
yev, 1988). This zoned alteration mineralogy, in many of the southwest Pacific porphyry
copper systems, grades from an inner potassic zone outward to progressively cooler
propylitic alteration assemblages, (e.g. Yandera, Watmuff, 1978; Panguna, Ford, 1978; Fig.
5.5; also for El'Salvador, Gustafson and Hunt, 1975). A similar zonation in altera-tion
mineralogy is associated with the initial emplacement of high level intrusions in active
Philippine geothermal systems (chapter 2).
In calc- alkaline magmatic arc systems, the potassic zone is dominated by biotite. The
biotite, in some systems (e.g., Yandera), varies from brown to green brown from the core
to the margin of the potassic alteration, which is a reflection of an increase in Mg/Fe
content (Watmuff, 1978). In areas of crustal rifting and more felsic/silicic intrusions, the
potassic zone in the intrusion is dominated by K-feldspar (e.g. Goonumbla, Heithersay and
Walshe, 1996; Cadia, Newcrest, 1996; Dinkidi, Garrett, 1996). In both environ-ments,
secondary quartz and plagioclase are subordinate to the biotite. Secondary clinopy-roxene
is intergrown with biotite in the central potassic zones in the St. Thomas II porphyry
copper deposit (Coote, 1993) and the Palinpinon active porphyry copper sys-tem,
Philippines (Reyes, 1990). Magnetite is almost ubiquitous in the potassic zone, and is
interpreted to reflect neutral to alkaline, reducing conditions.
The early potassic zone alteration commonly exhibits hornfelsic textures (McMillan and
Panteleyev, 1988) and have a composition similar to that of primary magmatic biotites
(Watmuff, 1978). Isotopic studies (Sheppard et al., 1971; Ford and Green, 1977) illus-trate
that the biotite is formed from magmatic fluids, and may have formed in response to
reaction between the melt and primary hornblende (Burnham, 1979). Later biotite occurs
in veinlets and is more Mg-rich (Watmuff, 1978). This change in the style and composi-
tion of alteration is here inferred to indicate that there is a gradation from heat being trans-
ferred from the melt initially through conduction (contact metamorphic), to

64
Exploration Workshop "Southwest Pacific rim gold-copper systems:Struciure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

later episodes where heat commences to be transferred through hydrothermal

convection, possibly associated with the early exsolution of magmatic fluids.
Alteration zones distal to the central potassic core reflect progressively cooler conditions
(Fig. 5.5). Secondary amphiboles, typically actinolite in calc-alkaline environments, lo-
cally occur with biotite in the outer potassic zones, and with epidote in the inner propylitic
alteration zones (e.g. Philippine systems, Sillitoe and Gappe, 1984; Mamut, Kosaka and
Wakita, 1978; Grasberg, MacDonald and Arnold, 1994). At shallower levels, propylitic
assemblages of epidote-chlorite-albite-carbonate grade upward and outward into zones
dominated by chlorite and progressively more hydrated zeolites which form under low
temperature conditions (e.g. Yandera; Watmuff, 1978). Prehnite (e.g. Cadia, Leach, un-
published data), and very rarely pumpellyite (Leach, unpublished reports), are encoun-
tered in some epidote-bearing propylitic assemblages. Although magnetite is locally
abundant in the propylitic zones, pyrrhotite (e.g. Mamut, Kosaka and Wakita, 1978) and/or
pyrite dominate in these cooler, less reducing, and locally lower pH environments. In
places magnetite and/or pyrrhotite produce significant geophysical magnetic anomalies
within the propylitic alteration zones (e.g., Batu Hijau; Meldrum et al., 1994),

In the central portions of many of the intrusions, potassic alteration grades with
increas-ing depth through zones of propylitic alteration (commonly with actinolite) to
zones of in-creasingly weaker alteration, and in some cases to a relatively unaltered
core (e.g. Dinkidi, Garrett, 1996). Elsewhere, the deep propylitic alteration assemblage
is dominated by al-bite (e.g., Yandera, Watmuff, 1978; Batu Hijau, Indonesia; Meldrum
et al., 1994). This albitisation of plagioclase has been interpreted to have formed during
the late stage crys-tallisation of the melt (Watmuff, 1978), since all other alteration
events post-date the al-bite. Albitisation of plagioclase at deep levels in porphyry
systems has also been docu-mented at the Ann-Mason deposit, Nevada, although there
it is late in the sequence of events and is inferred to be related to the incursion of non-
magmatic fluids (Dilles and Einaudi, 1992).

Stage II: Stockwork and Sheeted Quartz Veins

The cooling and crystallisation of the melt results in fracturing, especially around the car-
apace of the intrusion, and this fracturing is accompanied by the exsolution of magmatic
volatiles (Henley and McNabb, 1978). Pressure within the cooling magma may have local-
ly exceeded lithostatic pressure and this may initiate fracturing in the brittle host rocks
(Burnham, 1979). Tectonic movements are inferred to have facilitated this fracturing of the
carapace and may have locally promoted a sudden pressure drop (section 3.viii). As
outlined in section 4, a change from lithostatic to hydrostatic pressures at 2 km depth, and
an accompanying drop in temperature upon boiling, can result in the deposition of quartz
from the magmatic fluids into the fractured carapace.
Quartz can therefore be deposited within the fracture network and results in the
formation of a stockwork quartz vein system around the carapace of the intrusion
(Figure 5.5). However as outlined in section 3.vii, the quartz is also commonly
deposited in sheeted fracture systems, typically at the fault controlled margins of the
intrusion (e.g. Mamut, Kosaka and Wahila, 1978). In some cases the sheeted quartz
veins crosscut the host in-trusion and extend up to a few kilometers into the host
sediments or volcanics (e.g., Ca-dia; Newcrest, 1996). Elsewhere the development of
these sheeted veins are interpreted to relate to continued reopening of the dilational
faults which facilitated the initial em-placement of the intrusion into shallow crustal
levels (e.g., Frieda River; Leach, un-published reports). These sheeted plumbing
systems are here inferred to be important in subsequent metal transport.
This Stage II quartz is deposited from a hot, hypersaline two phase fluid. The quartz is
commonly a murky light grey-white color due to the large number of both primary and
secondary fluid inclusions and minute inclusions of gangue and ore phases. This early

65
Exploration Workshop "Southwest Pacific rim gold-copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

quartz is distinguishable from the clearer and whiter quartz veins formed under cooler,
more dilute conditions (Stage II) and which contain fewer fluid and mineral inclusions.

At Batu Hijau, average fluid inclusion homogenisation temperatures in the quartz are
highest towards the carapace of the mineralized tonalite, and decrease both with depth
and at shallower levels (Coote, 1992). This distribution implies that fracturing and
quartz vein development initially took place around the upper margins of the intrusive,
then later ex-tended out into the country rock, and inward to the core of the intrusion.
In some systems, the stockwork quartz veins near the carapace of the intrusion locally
comprise >20-30 percent of the intrusion (e.g., Grasberg, MacDonald and Arnold,
1994; Batu Hijau, Irianto and Clarke, 1995). This high density of quartz veins implies
that a significant abundance of volatiles were released from the melt. In these cases
the inter-preted original enrichment in volatiles, possibly during melt ascent
(Lowenstern, 1994), may have provided the buoyancy necessary to facilitate
emplacement at relatively shallow crustal levels.
Two stages of early quartz veining have been recognised by a number of authors in the
southwestern USA porphyry copper systems (Nash, 1976; Reynolds and Beane, 1985),
and these also occur in porphyry copper systems in the southwest Pacific. Early quartz
veins at Goonumbla (Heithersay and Walshe, 1996) formed during late stages of melt
crystallisation as evidenced by their local discontinuous ptygmatic morphology which
merge with the host intrusion. These are equivalent to the 'A' -type quartz veins of Gus-
tafson and Hunt (1975). Fluid inclusion analyses at Panguna and Frieda River (Eastoe
1978) as well as at Goonumbla (Heithersay and Walshe, 1996), indicate that this early
veining formed at temperatures of >6OO-8OO°C, similar to crystallisation temperatures
of the melt, and from brines with salinities of >35-40 weight percent NaCl.
The vast majority of the quartz veins commonly exhibit multiple phases of fracturing and
sealing within a brittle, fractured, host intrusion (e.g. Panguna, Eastoe, 1978), and there-
fore post-date crystallization of the melt at that level. They also crosscut the earlier formed
potassic and inner propylitic alteration zones, and may be accompanied by K-feldspar
alteration of the adjacent wallrock plagioclase of Stage I biotite (Watmuff, 1978; Fig.
5.8).These are equivalent to the 'B'-type quartz veins of Gustafson and Hunt. Mag-netite,
K-feldspar, biotite and/or albite are locally intergrown with the quartz, and may form as
vein selvages to quartz stockwork veins (e.g. Goonumbla, Heithersay and Wal-she, 1995).
Minute biotite and anhydrite inclusions are also encountered in some quartz stockwork
veins (Britten, 1981). In systems associated with alkali intrusions, K-feldspar and/or
magnetite are significant components in both the stockwork and the sheeted veins (e.g.
Goonumbla, Heithersay and Walshe, 1995; Dinkidi, Garrett, 1996).

Fluid inclusion studies (e.g., Eastoe, 1978; Heithersay and Walshe, 1995), indicate that
the majority of the stockwork and sheeted quartz veins were typically deposited from a
hot (> 300-500°C), hypersaline (> 25-30 wt percent equivalent. NaCl), two phase (i.e.,
boiling) brine. Recent work from porphyry systems in the southwest USA (e.g.., Cline
and Bodnar, 1994) has shown that during the formation of such quartz veins, the vapour
and brine fluids partition separately from the cooling melt. Therefore, the salinity of the
liquid-rich brine, as determined from fluid inclusion data, may be an over-estimation of
the true salinity of the fluids which exsolved from the melt.
It is evident that the fluids which formed the stockwork quartz veins were significantly
enriched in metals. Hematite, magnetite and copper minerals commonly occur as minute
phases in the primary saline-rich fluid inclusions (Roedder, 1984). Eastoe (1978) esti-
mated a Cu concentration of l,900ppm from one liquid inclusion in quartz veining at
Panguna. Similar concentrations of copper (average = 2,000ppm Cu), in addition to high
concentrations of other metals (up to 35,000 ppm Fe, 2700 ppm Zn, and 940 ppm Pb)
have been detected in hypersaline brines in primary inclusions in the Questa porphyry
molybdenum system, southwest USA (Cline and Vanko, 1995).

66
Exploration Workshop "Southwest Pacific rim gold-copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

However, experimental work (Hemley et al., 1992) has illustrated that solutions at tem-
peratures of 500°C and lkbar pressure are as saturated at similar concentrations
(average = 1300ppm Cu) as the fluids which deposited the sheeted and stockwork
quartz veins, but at much lower salinities (5.8 weight percent equivalent NaCl). Since
the solubility of copper as a chloride complex increases with increasing salinity (Crerar
and Barnes, 1976), the hypersaline (>25-30 weight percent NaCl) solutions from which
the quartz veins were deposited were therefore probably significantly undersaturated
with respect to copper (Roedder, 1984).
This experimental work is supported by detailed petrology on porphyry copper systems in the
southwest US (Beane and Titley, 1981; Reynolds and Beane(1985), which indicates that copper
mineralisation is not associated with deposition of the quartz veins from these hot (>350-
400°C) hypersaline magmatic fluids. This is also the case for southwest Pacific porphyry
copper systems (Leach, unpublished reports). Quartz-magnetite stockwork veins at Yandera
form a barren core, whereas mineralisation is here associated with later structur-ally controlled
sericite and zeolite veining (Titley et al, 1978; Watmuff, 1978). Early quartz veins at Frieda
River, which were deposited at temperatures of >400-500°C from hyper-saline brines, are
barren and merely provide a brittle host for later mineralization (Leach, unpublished data).
Similarly, copper mineralisation at Copper Hill, NSW, is associated with late sericite-chlorite
veins which crosscuts the stockwork quartz veins (Scott, 1978).

Therefore, although the early hypersaline fluids contained ore metals when they exsolved
from the crystallising magma (Bodnar, 1995), these fluids did not deposit copper (or
gold) during formation of the quartz veins. However, the high concentrations of iron in
both fluid inclusions (Cline and Vanko, 1995) and high temperature, saline solutions
(Hemely et al., 1992), implies that these brines were near saturation with respect to iron
oxides as they exsolved from the cooling intrusion. This is supported by the abundance
of magnetite found associated with the formation of early potassic-propylitic alteration,
and with later quartz and K-feldspar stockwork and sheeted veins.
Magmatic fluids outflow laterally and vertically along regional fracture/fault systems
(Henley and McNabb, 1978). It is proposed here that these magmatic fluids are
entrained into circulating waters and deposit quartz and K-feldspar/adularia in veins
which are a host to later gold and base metal mineralization (chapter 7).
Advanced Argillic Alteration
The intense silicification and advanced argillic alteration along the upper margins of some
southwest Pacific porphyry systems are here interpreted to have formed during the exso-
lution of magmatic volatiles from the crystallising high level stock (Figure 5.5). These
zones of advanced argillic alteration have been documented in a number of southwest Pa-
cific porphyry copper systems; e.g., at Batu Hijau (Meldrum et al, 1994), Horse Ivaal,
Frieda River (Britten, 1991), Dizon, Philippines (Sillitoe and Gappe, 1984), Cabang Kiri
(Carlile and Kirkegaard, 1985), and Lookout Rocks, New Zealand, (section 6.ii.b.l).

Advanced argillic alteration, which is distributed along the margins of the mineralised in-
trusions, is strongly aligned within bounding structures. Evidence from the Palinpinon ac-
tive porphyry system, and the solfataras at Biliran (Mitchell and Leach, 1991) indicate that
the silicification and advanced argillic alteration may extend from porphyry depths to the
surface where they manifest as magmatic solfataras. At Palinpinon, the advanced argillic
alteration post dates the formation of zoned potassic-propylitic and skarn assemblages, but
pre-dates later phyllic and argillic alteration. In the Alto Peak geothermal field, the acidic
alteration has been shown to relate to a hot (<400°C) magmatic-dominated vapour-plume,
which is inferred to be currently exsolving from a

67
Exploration Workshop "Southwest Pacific rim gold -copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

crystallising magma at depth, and overprints earlier potassic-propylitic alteration

(Reyes et al., 1993).
Alteration in these advanced argillic zones marginal to high level porphyry intrusions is
zoned from: biotite in the potassic zone, through zones of K-feldspar and chlorite,
sericite-andalusite ± corundum ± tourmaline, pyrophyllite-diaspore, to alunite +
kaolinite (e.g. Cabang Kiri, Lowder and Dow, 1978; Frieda River, Britten, 1981, Leach,
unpubl. data). This zonation is interpreted to reflect an increase in fluid acidity as
reactive magmatic vola-tiles (mainly SO2 and HC1, and subordinate but locally
significant HF) disproportionate and dissociate at progressively cooler conditions away
from the crystallizing melt (Rye et al., 1992). These features are discussed in more detail
in the high sulphidation section (Section 6).

Stage III: Cooling and Metal Deposition

As discussed above, copper-gold mineralisation in many porphyry copper systems in
both the southwest USA (e.g. Beane and Titley, 1981; Reynolds and Beane, 1985; Dilles
and Einaudi, 1992) and the southwest Pacific (e.g. Watmuff, 1978; Holtzberger et al.,
1996), post -date Stage I potassic/propylitic alteration, quartz veining and advanced
argillic altera-tion (Figure 5.6). Thus mineralization is attributed by many authors to be
a result of a pro-gressive change from an environment dominated by magmatic fluids to
one that is domi-nated by cooler and more dilute meteoric waters (e.g., Gustafson and
Hunt, 1975; Reyn-olds and Beane, 1985).
The general sequence of events associated with Stage III cooling and metal deposition in
southwest Pacific porphyry systems is illustrated in Figure 5.9 and may be summarised
as: quartz --> silicate mineral (potassic, propylitic, phyllic) —> Fe-oxide/sulphide —>
mineralisation (Cu-Au ± Mo, Pb, Zn) --> sulphate / carbonate. These minerals occur :
i) in newly formed fractures which crosscut Stage I alteration and quartz veining
ii) in older reopened fractures
iii) in open spaces in Stage I quartz veins (commonly along partings in the
central of the veins)
iv) associated with the alteration of primary and pre-existing secondary
feldspars and mafics in the wallrock.
The quartz is commonly clear to white, and from fluid inclusion analyses (Eastoe
1978; Watmuff, 1978) was deposited from a fluid which was significantly more
dilute (<15 weight percent NaCl, and typically <5 weight percent NaCl) and cooler
(<300-400°C) than the fluid that deposited Stage I quartz. Where these quartz veins
are associated with sericite-chlorite assemblages they are equivalent to the 'D'-type
veins of Gustafson and Hunt (1975)
The silicate phases are intergrown with, or commonly overgrow the quartz, and are
zoned from early and deep to shallow and late as : potassic minerals (biotite then K-
feldspar) --> calc-silicate minerals (actinolite, epidote and zeolites) —> chlorite -->
sericite --> illitic clay --> kaolin clay. Some or all these may be present in any one
system. Mineralization is intimately associated with these silicate phases, although the
ore phases commonly overgrow the silicates (Leach, unpublished data). This spatial
and temporal zonation is indicative of progressive cooling and decrease in the pH of the
fluids during mineraliza-tion (Figure 5.8; Beane and Titley, 1981).
Biotite and K-feldspar are early Stage II minerals which were deposited in thin veinlets
(generally <5mm in width) with quartz, magnetite and/or pyrite, and are associated with
variable abundances of chalcopyrite (e.g., Yandera, Titley et al., 1978). The occurrence of
biotite indicates that the veins and alteration formed under neutral to alkaline conditions at
temperatures of >3OO-35O°C (section 4.ii.f). Although K-feldspar is locally intergrown
with the biotite, it may also occur in thin veinlets which cut the biotite (Titley et al.,

68
Spacial and temporal distribution of minerals during Cu-Au mineralization in Southwest Pacific porphyry copper systems.
Exploration Workshop "Southwest Pacific rim gold-copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

1978). The magnetite, which is intergrown with, but commonly overgrows the biotite, may
contain inclusions of chalcopyrite and can therefore be distinguished from magnetite that
formed during Stage I potassic alteration and igneous magnetite (e.g., Taysan, Leach,
unpubl. data). Similar K-feldspar-quartz-magnetite-chalcopyrite veinlets in southwest USA
porphyry copper deposits were deposited from less saline fluids (<15 weight percent NaCl)
and under cooler conditions (200-450°C) than those for stockwork quartz veins. Iso-topic
data has shown that these minerals were deposited from mixed magmatic-meteoric fluids
(Beane andTitley, 1981; Sheppard, 1971; Reynolds and Beane, 1985).

The calc-silicate mineral phases (actinolite, epidote and zeolites) occur in fractures and in
open spaces, replace wallrock minerals (mainly mafics), and post-date the biotite and K-
feldspar (Sillitoe and Gappe, 1985; Chivas, 1978; Watmuff, 1978). The paragenetic se-
quence of actinolite --> epidote -- > zeolites (e.g. Cadia, Leach unpubl. data; Taysan,
Leach, unpubl. data) is indicative of progressively cooling conditions under near neutral
fluid pH conditions (Fig 5.8). The zeolite phase is typically laumontite (e.g. Mamut,
Kosaka and Whila, 1978; Cadia, Leach, unpubl. data). In some systems prehnite is early
and occurs at deeper levels than the laumontite (e.g. Cadia, Leach, unpubl. data). Else-
where more hydrated and lower temperatures zeolites such as stibnite (Yandera, Watmuff,
1978) and chabazite (Panguna, Eastoe, 1978) are late, post-mineral and associated with
barren calcite veining. Chlorite is typically associated with the above calc-silicates and in
many cases replaces early Stage I & II biotite (Sillitoe and Gappe, 1984).
Magnetite, locally with chalcopyrite inclusions, and/or pyrrhotite are in places
associated with early actinolite and epidote deposition, however pyrite generally
dominates the iron minerals in the calc-silicate assemblages (Watmuff, 1978; Chivas,
1978; Leach, unpubl. data). The association of actinolite with copper minerals
indicates mineralization occurred at temperatures >280-300°C. Hematite alteration of
magnetite is inferred have occurred during chlorite alteration of biotite at Taysan
(Leach, unpubl. data) and Frieda River (Leach, unpubl. data).
Chalcopyrite and minor bornite commonly are associated with the calc-silicate minerals.
In many Philippine porphyry copper systems (Sillitoe and Gappe, 1984), at Yandera
(Watmuff, 1978) and some southwestern USA deposits (e.g. Ann-Mason, Nevada, Dilles
and Einaudi, 1992), the bulk of the copper mineralization is associated with the late stage
chlorite-epidote phase of veining and wallrock alteration. The association of epidote and
laumontite with the copper sulphides indicates mineralisation took place under near
neutral fluid pH at temperatures of 15O-3OO°C (section 4.ii.f)-
Most of the copper-gold mineralisation in the southwest Pacific systems is intimately as-
sociated with late chlorite (e.g. Batu Hijau, Irianto and Clark, 1995) and/or sericite or il-
litic clay (e.g. Copper Hill, Scott, 1978; North Sulawesi, Lowder and Dow, 1978; FSE,
Garcia, 1991; Frieda River, Leach, unpubl. data) deposition and wallrock alteration.
Copper- gold mineralisation is also predominantly associated with the sericite -chlorite
event in porphyry copper deposits in the southwest USA (Beane and Titley, 1981). In
this phase of deposition/alteration, chlorite dominates at depth and is early, whereas
sericite dominates at shallower levels and is late (e.g. Frieda River, Leach, unpubl.
data). The upwards zonation of chlorite to sericite is indicative of a progressive
decrease in fluid pH af shallower levels. The change from calc-silicate minerals to
chlorite and then to sericite also reflects a decrease in fluid pH during progressively
later stages of mineralization (Fig 5.8). Isotopic analyses indicate, that the sericite in
many porphyry systems (Sheppard et al, 197'ifFord' and Green, 1977; Eastoe 1978) is
derived from meteoric dominated wa-ters. However, Wolfe (1994) interpreted from
isotope analyses, that the sericite associated with mineralisation at the E48 stock at
Goonumbla was derived from magmatic-dominated fluids, whereas post-mineral
sericite was probably formed from a meteoric-dominated water.

69
Exploration Workshop "Southwest Pacific rim gold-copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

Chalcopyrite is more abundant than bornite in chlorite-dominated assemblages, whereas

bornite is locally more abundant than chalcopyrite in sericitic assemblages. Bornite is
commonly intergrown with, and locally overgrows chalcopyrite. Phases of the
intermedi-ate solid solution series (ISS, e.g. idaite) are rare and commonly late, possibly
formed un-der lower temperature conditions (e.g. Wafi River, Leach, unpubl. data).
Hypogene chal-cocite, covellite, enargite and tennantite generally post-date the bornite,
and are restricted to sericitic assemblages at shallow levels in some systems (e.g. Cadia,
Leach, unpubl data; Goonumbla, Wolfe, 1994; Frieda River, Leach, unpubl. data), and in
peripheral zones of pyrophyllite and diaspore at Dizon (Malihan, 1987).
Molybdenite is generally associated with chlorite-epidote-carbonate deposition and
altera-tion (Watmuff, 1978; Sillitoe and Gappe, 1984). Galena and sphalerite are
typically very late and are commonly associated with sericite (e.g. Goonumbla, Wolfe,
1994) and car-bonate (e.g. Copper Hill, Scott, 1978) veins and shears which are post-
copper mineralisa-tion.
Gold as the native metal, typically occurs as minute (<10-15u,) inclusions in the copper
sulphides. More gold occurs with bornite than chalcopyrite (Sillitoe and Gappe, 1984)
and is in places enriched in hypogene covellite and chalcocite (e.g. Goonumbla; Wolfe,
1994). It is also more commonly associated with.sericite alteration and related veins
rather than with chlorite, calc-sdifcate or potassic assemblages (Leach unpubl. report).
Gold generally has" a high fineness (>900, Figure 4.8), although where cooler
conditions are evident (e.g., locally at Goonumbla; Leach, unpubl. report) the fineness
is lower and more comparable to that encountered in mesothermal vein systems. High
gold grades commonly, but not ubiquitously, occur at shallow levels (e.g. Ok Tedi, Rush
and Seegers, 1990) or in fault controlled porphyry margins (e.g. Horse Ivaal, Leach,
unpubl. report; Didipio, Garrett, 1996) and is here associated with extensive sericite
alteration and deposition. High grade gold at Copper Hill is associated with late
carbonate-base metal veins, which crosscut porphyry-related quartz veins and pyrite-
chalcopyrite-chlorite/sericite veins (Leach, un-publ. report; section 7).
Anhydrite, calcite and locally dolomite commonly overgrow the sulphides in
veins/veinlets and wallrock alteration (Leach, unpublished data), although in some
pro-spects the sulphide mineralisation extends into the sulphate and carbonate event
(e.g., Frieda River, Leach, unpubl. report). Fluid inclusion data from Frieda River
(Eastoe 1979; Leach, unpubl. report) indicate that quartz and anhydrite, which is
associated with copper mineralization, was deposited from fluids at significantly lower
(100-400°C; aver-age = 230°C) temperatures and which were more dilute (<20-25
weight percent NaCl) than those that deposited the Stage I quartz (Th average =
440°C; salinity >35-40 weight percent NaCl). Isotope analyses indicate that sulphides
and anhydrites from Panguna and Frieda River (Eastoe 1983) are in isotopic
equilibrium and that the sulphur in the anhy-drite is of probable magmatic origin. It is
suspected here that the carbonates in southwest Pacific porphyry systems were
deposited in response to mixing of magmatic- and meteor-ic-derived fluids. This is
supported by isotopic studies by Zaluski et al. (1994) on calcite associated with sericite
and copper mineralization in the Babine porphyry copper deposit, British Columbia.

Stage IV : Phyllic. Argillic and Advanced Argillic Overprint

Pervasive quartz-sericite-pyrite alteration which has an increasing abundance of
chlorite at depth, and which post-dates copper-gold mineralisation, overprints other
alteration as-semblages and veins at shallow levels and along the intrusive margins of
a number of southwest porphyry copper systems (e.g., Yandera, Watmuff, 1978; Frieda
River, Brit-ten, 1981; Batu Hijau, Meldrum et al., 1994; Goonumbla, Heithersay and
Walshe, 1996). Isotopic data from southwest Pacific (e.g. Ford and Green, 1977;
Wolfe, 1994; Heithersay and Walshe, 1996) indicates that this sericite is derived from
a meteoric-dominated fluid.

70
Exploration Workshop "Southwest Pacific rim gold-copper systems:Siructure. Alteration and Mineralization: Corbelt GJ & Leach TM. S/96

Argillic alteration overprints the sericite-chlorite alteration in a number of porphyry cop-

per deposits (e.g., Batu Hijau, Meldrum et al., 1994; Grasberg, MacDonald and Arnold,
1994; Frieda River, Leach unpubl. report) and is equivalent to the clay in the sericite-clay-
chlorite zones at shallow levels in many Philippine systems (Sillitoe and Gappe, 1984).
Clays at Yandera are zoned from illite in the central part of the deposit, to illite-smectite in
outer zones (Watmuff, 1978). Late quartz-clay-chlorite veinlets cut sulphide veinlets at
Yandera, although some may be related to late copper mineralisation (Watmuff, 1978).
High grade Cu-Au mineralisation in late chalcopyrite-filled fractures at Grasberg is
intimately associated with illitic clay deposition (Leach, unpublished report). Smectite
and/or kaolinite ± siderite alteration and deposition in open spaces and fractures in many of
the porphyry copper systems (e.g. Cadia and Taysan, Leach, unpubl. report) are in-ferred,
from x-ray diffraction data to be late hydrothermal, although in some cases these clays are
of supergene origin. Isotope studies on late argillic alteration in the Babine porphyry
copper deposit, B.C. (Zaluski et al., 1994) indicates that the clay there is de-rived from
rock reaction with meteoric waters within a circulating hydrothermal system.

In some porphyry copper systems, renewed igneous activity has taken place during the
later stage of meteoric water incursion (Sillitoe and Gappe, 1984). The emplacement of
late stage stocks has locally resulted in the formation of diatreme breccia complexes prox-
imal to the mineralised porphyry, and these may post-date and cross-cut copper-gold
mineralization (e.g. Dizon; Malihan, 1987). In some circumstances, exsolution of mag-
matic volatiles from these late stocks may have resulted in overprinting of the mineral-ized
porphyry by hot acidic fluids (e.g., Wafi River, Erceg et al, 1991). Elsewhere, the
advanced argillic alteration has been late and metal destructive (e.g. El Salvador, Chile,
Gustafson and Hunt, 1975; Island Copper, Canada, Mathias et al., 1995). The relation-ship
between advanced argillic alteration and copper-gold mineralisation is discussed in more
detail in the following chapter on high sulphidation systems.

d) Environments for Porphyry Copper-Gold Mineralization

The emplacement of melts at shallow crustal levels is inferred to occur along regional di-
lational structures and is here interpreted to produce initial zoned potassic-propylitic al-
teration within a magmatic-dominated regime. Hypersaline brines ( > 25-30 weight per-
cent NaCl) and volatiles were released from the melt as the upper levels cooled and crys-
tallised, and deposited quartz and/or K-feldspar within stockwork and sheeted fracture
systems at temperatures of >400-600°C. The release of the fluids from the melt may have
been facilitated by the reactivation of the dilational structures through tectonic movement.
The cooling of these fluids at shallower levels is inferred to have resulted in the dissocia-
tion of dissolved magmatic volatiles and the progressive formation of hot acidic fluids
(Rye et al., 1992), and subsequent advanced argillic alteration through rock reaction.
These events are postulated to be periods of exsolution of metals from the melt, however
it is considered that the intrusion and immediate host rocks at this time were too hot, and
the fluids too saline, to provide an environment for metal deposition.
Copper-gold mineralisation in porphyry environments is indicated to have taken place
at temperatures of around 2OO-35O°C. Metal deposition is preceded by potassic and
calc-silicate and Fe -oxide/sulphide mineral deposition and alteration, and is
overgrown by later anhydrite and calcite/dolomite. These minerals infill pre-existing
fractures/veins, open cavities and vein partings, new fracture sets or is associated with
wallrock altera-tion. The zonation from early to late and deep to shallow of the silicate
minerals of: biotite --> K-feldspar --> actinolite --> epidote -- > zeolites --> chlorite
--> sericite --> pyro-phyllite --> kaolin/illitic clay, is indicative of progressive cooling
and decrease in fluid pH during copper-gold mineralization.

71
Exploration Workshop "Southwest Pacific rim gold-copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

This cooling may have taken place solely through heat conduction to the country rock in
small mineralised intrusions (e.g. Goonumbla). However in active porphyry copper sys-
tems in the Philippines, CO2-rich and dilute groundwaters have been encountered down to
depths of up to 1.5-2 km from the surface. These waters have reacted with the host rocks to
form similar zoned phyllic and argillic alteration (chapter 2) as described above for
porphyry copper systems. The model of an incursion of surfical waters to facilitate the
cooling of the upper levels of a larger mineralized porphyry intrusion is also indicated by a
meteoric isotopic signature in sericite, and the dilute conditions from fluid inclusion data.
The information from active hydrothermal systems suggests that the meteoric waters may
have migrated down the same structures as initially facilitated the emplacement of the in-
trusion to shallow levels (e.g. Bacon Manito geothermal field). It is speculated that this can
only take place once the intrusion has already cooled significantly. Pressure draw downs
along these structures may have been initiated by renewed intrusion elsewhere within the
immediate vicinity (e.g. from Cawayan to Pangas-Pulog, in the Bacon-Manito Geother-mal
Field, Philippines; section 2.iv.a.3.ii). Other authors (e.g. Gustafson and Hunt, 1975) and
computer modelling (Norton and Knight, 1977) suggest that the meteoric waters may have
been sourced from the margins of the intrusion.
Copper-gold mineralisation apparently takes place at some significant time period after
the intrusion has cooled and crystallised. K/Ar age dating at FSE, Philippines, has
indicated the time span between early biotite alteration and late illite associated with
copper mineral-isation may have been up to 200,000-300,000 years (Arribas et al.,
1995). It is therefore speculated that the magmatic fluids and metals associated with
mineralisation in a porphyry copper system have probably been exsolved from the
cooling and crystallizing of deeper melts of the same shallow level intrusion, or of a
much larger parent melt (Fig-ure 5.6).
The metal-bearing magmatic fluids are therefore interpreted to have migrated from the deeper
melts along reactivated fractures at the margin of the intrusion. Metal deposition takes place
as these fluids enter environments which have cooled to <3OO-35O°C, and in many cases
enter fractures, at these shallow levels, which may be saturated with cool and dilute meteoric-
derived waters. The cooling of these fluids results in progressive decrease in fluid pH in
response to the dissociation of dissolved gases. This change in fluid pH is reflected in the
change in the silicate species of: potassic/calc-silicate minerals --> chlorite
—> sericite —> pyrophyllite/kaolinite. The decrease in fluid pH may also have been
facili-tated-by the mixing of low pH CC>2-rich waters.
Southwest Pacific porphyry copper deposits are typically gold -rich (Sillitoe, 1993a).
Variations in the Au:Cu ratios of the porphyry copper systems are here interpreted to
re-flect, in part, a range from hotter environments of mineralization (more copper-
rich) as-sociated with potassic and calc-silicate assemblages (e.g., Yandera, PNG) to
those at cooler, more mesothermal and meteoric environments (gold-rich, e.g., Dizon
and Didip-io, Philippines) associated with sericite and/or chlorite assemblages.

ii) Skarn Deposits

a) Introduction
Skarns are rocks consisting of Ca-Fe-Mg-Mn silicates formed by the replacement of
car-bonate-bearing rocks during regional or contact metamorphism and
metasomatism (Einaudi et al., 1981) in response to the emplacement of intrusions of
varying composi-tions. Skarns can therefore be regarded as a specific type of
aleration within a porphyry environment.

72
Exploration Workshop "Southwest Pacific rim gold-copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

The terms exoskarns and endoskarns are used to describe deposits from sedimentary
and igneous/intrusive protoliths respectively. Veins of skarn mineralogy may be present
in both intrusions and carbonate sediments. Calcic skarns form by replacement of
limestone and produce Ca-rich alteration products such are garnets (grossular-
andradite) - clinopyroxene (diopside- hedenbergite), vesuvianite, and wollastonite.
Magnesian skarns form by the replacement of dolomite, and produce Mg-rich
alteration phases such as diop-side, forsterite and phlogopite. Magnetite is common in
magnesian skarns since iron is not taken up by the Mg-rich silicates.
Skarns typically have complex mineral assemblages and are polyphasal, with early
stages formed at high temperatures which creates assemblages of anhydrous silicates +
iron ox-ides. These are overprinted by later hydrous silicates and sulphides which are
formed at lower temperatures. Spatial mineralogical zoning is related to both lateral
and vertical dis-tance from the intrusion (i.e., to chemical potential and temperature
gradients) and to depth (i.e., to these gradients plus pressure; Meinert, 1993).
Detailed mapping of the distribution of alteration and ore phases provides information
about the overall size, characteristics and genesis of a skarn system, and these may
pro-vide vectors to help target exploration. Models of skarn zonation are particularly
useful in evaluating incompletely exposed or inadequately explored skarn systems.
Skarn deposits are not common in the southwest Pacific region, although significant
cop-per-gold skarn ore bodies occur in the Guning Bijih District, Indonesia (Ertsberg,
GBT, IOZ, DOZ, DOM and Big Gossan; Mertig et al., 1994), Ok Tedi, PNG (Rush and
See-gers, 1990), and Red Dome, Eastern Australia (Ewers et al., 1990).
As skarns are a specific class of porphyry system, they exhibit the same processes of
formation described previously for porphyry copper deposits. However, because the
host rocks have a specific chemistry, these processes are manifest in a different
manner. The following discussions are a summation of the work by Meinert
(1989,1993), Einaudi (1982a, 1982b) and Einaudi et al.(1981).

b) Processes of skarn formation

Skarn evolution occurs in response to three main sequential processes: the prograde iso-
chemical, prograde metasomatic, and late stage retrograde events (Fig 5.10). The iso-
chemical skarn event is equivalent to the formation of zoned potassic-propylitic alteration
formed in response to the conductive transfer of heat in porphyry copper systems. The
metasomatic skarn event is comparable to the formation of quartz stockwork veining and
advanced argillic alteration during exsolution of magmatic fluids from the crystallising
porphyry stock. Retrograde skarns are analogous to the collapse of meteoric waters and
contemporaneous mineralization events outlined in Section 5.i.e.

1. Prograde Isochemical (metamorphic, contact metamorphic, calc-silicate

hornfels) Skarns:
Isochemical skarns form when intrusions are emplaced into calcareous sediments
with little or no introduction of chemical components. H2O is released from the
intrusion and CO2 from the calcareous sediments. The skarn development is
controlled predominantly by temperature and the composition and texture of the host
rock, within a predominantly conductive regime.
This contact metamorphism forms zoned thermal alteration aureoles consisting of Ca-Al
silicates/hornfels in calcareous shale or marl, Ca-Mg silicates in silty dolomites and calc-
silicate marble and/or wollastonite in limestone. Metamorphic minerals are generally fine
grained and the metamorphism is likely to be more extensive and/or higher grade around
a skarn formed at relatively greater depth than one formed at shallower levels. Isochemi-
cal skarns are characteristically confined to the host lithologies, and the bulk

73
Fig 5.10 Processes in the evolution of skarn deposits (adapted and modified from Meinert, 1993)
Exploration Workshop "Southwest Pacific rim gold-copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

compositions for any given rock type are identical for all alteration zones. These
skarns display a wide variety of mineralogy for a given number of elements. The
metamorphic stage of skarn development is essentially barren of ore mineralization
(Einaudi et al., 1981).

Zonations in mineralogy in response to decreasing temperature, and increasing concentra-

tions of CO2, (i.e. progressively away from the intrusive), can be generalised as follows:

in dolomite -
garnet ---- > pyroxene ---- > tremolite ---- > talc/phlogopite;
in limestone -
garnet -----> vesuvianite + wollastonite ----- > marble.
These changes reflect an increasing abundance of quartz + calcite, and an increase
in the hydration of mineralogy away from the source intrusion.
The Fe-content of garnets increase toward the intrusion, whereas the Fe:Mg ratio
of pyroxenes decrease. Garnet is therefore commonly dark red-brown proximal
to the intrusive, becoming lighter brown in more distal settings, and pale green
adjacent to fringe marbles (Meinert, 1993).
Reaction (also termed local exchange, bimetasomatic, or calc-silicate banded) skarns form
during the metamorphic event by the mass transfer of non-volatile components on a local
scale between adjacent lithologies. Skarnoids result from metamorphism of impure li-
thologies with some mass transfer by small-scale fluid movement (Meinert, 1993).

2. Prograde Metasomatic (infiltration, replacement) Skarns:

The formation of isochemical skarns is followed by the development of a metasomatic
or hydrothermal stage characterised by the exchange of H2O, silica, aluminium and
iron, which exsolve from the crystallizing intrusive, and CO2, calcium, and
magnesium which are derived from the calcareous sediments. The release of magmatic
fluids causes hydro-fracturing within the cooling pluton and previously formed
hornfels/isochemical skarn, and facilitates the ascent of magmatic-dominated fluids
along the intrusive contacts, frac-tures, fissures, faults, sedimentary contacts, pre-skarn
dykes and sills, and other permea-ble zones (Meinert, 1993).
Minerals formed during metasomatic processes overprints, and commonly replaces,
ear-lier metamorphic phases, and is characteristically coarser grained. Metasomatic
skams typically contain very few phases for the number of components (mono- or
bimineralic assemblages), with the composition of the alteration mineralogy not
reflecting the compo-sition or texture of the host lithologies.
Zonations in mineralogy are similar to those encountered in isochemical skarns.
Garnets and pyroxenes progressively become more iron-enriched and magnesium-
depleted with time. Lower temperature phases commonly overgrow and replace
minerals formed under earlier hotter regimes (e.g., pyroxene replacing garnet).
Einaudi et al., (1981) suggest that sulphide and oxide deposition commences during
the latter stages of metasomatic skarn development. Magnetite mineralization
dominates over sulphides, forming either by replacement of garnet or pyroxene at the
intrusive-skarn contact, or in outer zones at the marble-skarn contacts.

74
The influx of acid fluids may inhibit skarn formation in favour of the development of
massive pyrite-sulphide replacement bodies and breccia pipes (e.g., Brisbee). In this
case wholesale silicification has been superimposed onto earlier calc-silicate skarns.
3. Retrograde
The previously discussed skarns are commonly referred to as prograde skarns,
forming end- members of a continuum which shows a progressive transition from
early metamor-phic to late metasomatic dominated events. Retrograde skarns form
when temperatures decline and fluid compositions are dominated by meteoric waters,
especially where skarns formed at shallow crustal levels.
Retrograde alteration is characterised by the replacement of earlier prograde anhydrous
minerals by late stage hydrous phases such as epidote, amphiboles, chlorite and clays;
and this reflects the leaching of calcium, and introduction of volatiles. Unlike
metasomatic skarns, retrograde skarns have complex multiphase mineral assemblages.
Einaudi et al., (1981) list the following as typical retrograde alterations:

This is the main mineralization event. Sulphides and iron oxides occur as disseminations
in, and in veins which crosscut, prograde skarns, or as massive replacements of marble. In
the same manner outlined for porphyry copper systems, sulphide mineralization and
retrograde alteration in skarn deposits is typically structurally controlled and crosscuts
prograde skarns, in some cases extending beyond the skarns. Sulphide assemblages of
pyrite-chalcopyrite-magnetite occur proximal to intrusions, and in distal settings bornite-
chalcopyrite dominate. This reflects a decrease in total iron concentration during later
stages of skarn development. The sulphides are interpreted to have been deposited in re-
sponse to either decreasing temperatures, neutralization of the hydrothermal solution (es-
pecially at the marble contact), or changes in oxidation state of the fluids. The association
of most ore phases with late stage retrograde assemblages can be interpreted to either:

i) indicate that the prograde skarn is merely a reactive host rock for later
mineralising fluids which were derived from a deep parent melt, or
ii) indicate that there has been remobilization of sulphides which were deposited
during prograde events.
Elsewhere ore mineralization appears to post-date all skarn phases and this possibly
indicates that the sulphides were derived from a different or separate intrusion.

c) Skarn Ore Deposits

Ore deposits which are hosted in skarns are classified as skarn deposits. The
following most commonly used classification of skarn deposits is on the basis of
the dominant metal, i.e., Cu, Au, Pb-Zn, Fe, Mo, W and Sn (Einaudi et al., 1981;
Meinert, 1993; Einaudi, 1982a; Einaudi, 1982b).
Copper-gold skarns (e.g., Ertsberg, Ok Tedi) and gold (e.g., Red Dome) skarns are the
most economically significant skarn deposits in the southwest Pacific rim, and are asso-
ciated with shallow level calc-alkaline porphyritic intrusions. Copper skarns are typical-
ly dominated by andradite (Fe-rich) garnets, with massive garnet proximal to the intru-
sion, which grades outward via zones which contain an increasing abundance of
Exploration Workshop "Southwest Pacific rim gold-copper systemstStructure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

pyroxene (Fe-poor), to distal vesuvianite and/or wollastonite near the marble contact.
The garnet grades from red-brown, to light brown, to green, and yellow with
increasing dis-tance from the pluton. Chalcopyrite dominates mineralization close to
the porphyry, whereas bornite occurs in wollastonite zones near the marble contact.
Intense retrograde alteration is common and typically epidote-actinolite/tremolite
replaces prograde garnet. The presence of specular hematite may reflect a shallow
oxidising environment of for-mation.
Gold skarns (e.g., Red Dome) are associated with diorite-granodiorite plutons and
com-monly contain sub-economic Cu, Pb, and Zn. Potassium-feldspar, scapolite,
vesuvianite, apatite and Cl-rich amphiboles are common. Arsenopyrite and pyrrhotite
are the main sul-phide phases which indicate a reducing environment. Most of the gold
occurs as electrum in close association with bismuth and telluride minerals. Gold
skarns can form in distal portions of large skarn deposits, the proximal parts of which
commonly represent signifi-cant copper skarn deposits.
Lead-zinc skarns occur in distal settings relative to the source intrusions. They
commonly grade outward from zones rich in skarn minerals to zones in which the skarn
mineralogy is poorly developed. In places skarn mineralogy may be almost totally
absent. Almost all minerals in lead-zinc skarns are manganese-rich; the pyroxene:
garnet ratio and the manga-nese content of pyroxenes increase away from the intrusion.
These skarns are therefore closely related to the porphyry-related carbonate-base metal-
style gold systems outlined in section 6.ii.
Elsewhere in the world, iron skarns are the largest known skarn deposits and although
they are mined principally for their magnetite content, they contain subeconomic
amounts of Cu, Co, Ni, and Au, Some are these are transitional to copper skarns. Iron
skarns oc-cur in back-arc basins of island arcs where they are associated with iron-rich
diabase to dio-rite intrusions (Meinert, 1993).
Molybdenum and tin skarns are not seen in the southwest Pacific rim, and are found in
continental rift environments associated with leucocratic and high-silica granites
respec-tively. Tungsten skarns occur in deeply eroded calc-alkaline granodiorite to
quartz monzo-nite batholiths.

iii) Breccia-Hosted Gold Deposits

Gold-bearing magmatic hydrothermal breccias form in volcanoplutonic terrains and dis-
play characteristics indicative of a magmatic association. Deposits of this type generally
represent large tonnage low grade gold resources. Discrete breccia bodies include: in
eastern Australia, Kidston (Baker and Tullemans, 1990; Baker and Andrew, 1991) and Mt
Leyshon (Paull et al., 1990); in USA, Golden Sunlight (Porter and Ripley, 1985); and San
Cristobal, Chile (Corbett, unpublished, reports; Egert and Kaseneva, 1995). Sillitoe
(1991b) distinguishes breccias which are derived from a higher temperature magmatic
fluid of the Kidston and Golden Sunlight type, from phreatomagmatic (gas driven) di-
atreme breccias which are common within carbonate-base metal gold deposits described
in Section 7.iii (e.g., Montana Tunnels, USA, Sillitoe et al., 1985; Wau, PNG, Sillitoe et
al., 1984). Mineralization associated with the magmatic hydrothermal breccias described
above (Kidston, Mt Leyshon, San Cristobal) therefore corresponds to the deeper quartz-
sulphide gold + copper classification (Section 7.ii).
Magmatic hydrothermal breccias provide pre-mineral ground preparation overlying
porphyry environments from which mineralized fluids are channelled. Sheeted frac-
ture/vein systems commonly provide channelways for fluid transport. The style of
min-eralization within most magmatic hydrothermal breccia systems might best be
described as of the low sulphidation quartz-sulphide gold-type. Kidston is an
example of one of these, and is discussed in Section 7.ii.d.

75b
Exploration Workshop "Southwest Pacific rim gold-copper systems:Siructure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

iv) Porphyry and Alkaline Gold Deposits

Sillitoe (1979) predicted that a class of gold-rich porphyry copper deposits or porphyry
gold deposits would emerge, of which the Marte gold deposit is a good example (Vila
et al., 1991). Many of the deposits cited by Sillitoe (1979) occur in association with
alkaline volcanoplutonism and so were classed by Bonham (1988) as alkalic gold
deposits. This theme was extended by Rock et al. (1989), who applied the essentially
textural term of lamprophyre to group geochemically similar calc-alkaline rocks
occurring through a wide range of geological time, and suggested that these magmas
could display primary gold en-richments (Rock, 1991).
The identification of gold mineralization in the Tabar-Lihir-Tanga-Feni Island Chain in
Papua New Guinea (Moyle et al., 1990, 1991; Licence et al., 1987; Nord Resources
Pro-spectus), which Wallace et al. (1983) describe as shoshonitic, and the similarity to
host rocks at Emperor Gold mine (Anderson and Eaton, 1990; Eaton and Setterfield,
1993); Porgera (Richards, 1990) and Goonumbla, eastern Australia (Heithersay et al.,
1990), prompted the evaluation of potassium-rich rock types during the 1980's (Muller,
and Groves 1993, 1995).
The study of granite types evolved the classification of A-type granites (Collins et al.,
1982; Clements et al., 1986), which became popularly defined by explorationists as; "an-
hydrous, alkaline (potassium-rich), anorogenic, aluminous and anomalous", but promoted
some controversy in the application of mineral exploration (Hannah and Stein, 1990). In a
review, Pitcher (1993) suggests that the key factor in the mineralization of A-type granites
is the greater abundances of F, Cl and often B, and goes on to describe alkali fluoride
complexes as efficient means of transporting metals, most evident in tin systems. The high
temperature and fluxing effect of halogens aid in the transport of these phenocryst-poor
intrusions (Pitcher, 1993), commonly seen as dykes.
Recent models (Johnson, 1987; Solomon, 1990; Wyborn, 1992; Solomon and Groves,
1994) suggest that shoshonites are derived by the remelting of mantle derived material
and the arc reversal model of Solomon (1992) is consistent with the setting of
shoshonitic volcanism in the Tabar-Lihir -Tanga- Feni arc, PNG and Fiji. Miocene
volcanic arcs formed north of Papua New Guinea overly a south dipping subduction
zone (Fig. 1.2), which became clogged by the Pliocene collision of the Otong Java
Plateau. A new north dipping subduction subsequently developed south of New
Britain and remelting of al-ready subducted mantle material gave rise to the Pliocene-
Pleistocene Tabar-Lihir-Tanga-Feni Island Arc within NS trending rifts formed by the
arching of the subducting plate (Fig. 1.2).
It appears that shoshonitic magma types may preferentially give rise to gold and
copper deposits in particular tectonic settings. The dry and high temperature mantle-
derived melts must rise quickly from considerable depths and so commonly display an
association with major crustal structures or rifts. Shoshonite-related southwest Pacific
gold -copper depos-its occur in a range of low sulphidation intrusive-related settings
described in this manual as:
Porphyry Cu/Au- Goonumbla, eastern Australia; Marian, Didipio, Philippines
Quartz-sulphide Au - Lihir, Simberi in PNG Carbonate-base metal Au -
Porgera, PNG Epithermal Au/Ag - Emperor, Fiji
Thus the "alkaline gold deposits" are not a separate group of deposits, but are porphyry-
related gold systems which demonstrate an association with a similar, and possibly pro-
spective, magma source. Arribas (1995) notes that no high sulphidation copper-gold min-
eralization occurs in association with these intrusive compositions. In a comparison

75c
Exploration Workshop "Southwest Pacific rim gold-copper systems:Structure. Alteration and Mineralization: Corbett GJ & Leach TM. 8/96

of several alkaline gold deposits, Richards (1995) stresses that ore-forming processes
are common to many porphyry-related hydrothermal copper-gold systems, and
provides a model for possible mechanisms of concentration of chalcophile elements
in the magmatic volatile phase in alkaline systems, which illustrate typical zonations
from copper to gold-rich.

75d
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

6 HIGH SULPHIDATION GOLD-COPPER SYSTEMS

i) Characteristics

a) Introduction

High sulphidation gold-copper systems have also been termed acid sulphate (Hayba et al.,
1985) or alunite-kaolinite ± pyrophyllite (Berger and Henley, 1989) and included in the epi-
thermal class of gold deposits (Sillitoe 1993b, White and Hedenquist 1995). Bonham (1986,
1988) distinguished the high sulphidation style of gold deposits on the basis of:
* Abundance of sulphur as sulphate and sulphide,
* Zoned alteration as central advanced argillic, to argillic, to peripheral propylitic
alteration zones,
* Dominance of enargite/luzonite in the ore mineralogy,
* An association with calc-alkaline volcanism.

Early work by Urashima et al., (1981) recognised the alteration zonation at Iwato in the
Nansatsu deposits, while the alteration and ore mineralogy as well as the association with
porphyry copper systems are apparent in work of Sillitoe (1983).

The distinction between high and low sulphidation fluids is described in detail in Section l.iv.
and the characteristics of low and high sulphidation deposits in Table 3. High sulphidation al-
teration systems form as hot acid magmatic-derived fluids which are enriched in reactive vola-
tiles are cooled and neutralised by reaction with host rocks and groundwaters.

Although occurring outside the porphyry environment and hence commonly termed epither-mal,
high sulphidation alteration and mineralization also occur at crustal levels typified by
mesothermal porphyry deposits and so the term epithermal is avoided here. We suggest that the
term acid sulphate be utilised for alteration formed by collapsing low pH, surficial fluids dis-
cussed in Sections 1 and 4.

b) Classification

High sulphidation systems form at different crustal levels. The recognition of andalusite and
corundum in high sulphidation advanced argillic alteration (e.g., Horse-Ivaal, Frieda River,
PNG; Lookout Rocks, New Zealand; Cabang Kiri, Indonesia) suggests that some systems
formed under very hot conditions, at near-porphyry depths. Central alunite-pyrophyllite alter-
ation (e.g., Nena, Frieda River and Wafi River, PNG) are indicative of mesothermal to epi-
thermal conditions. The dominance of pyrophyllite over alunite (e.g, Pueblo Viejo, Dominican
Republic; Temora, Australia; Summitville, Goldfield and Red Mountain deposits in Western
USA), all point to deep to moderate epithermal levels of deposition. The occurrence of only
pyrophyllite and/or dickite/kaolinite and illitic clays in other systems (e.g., Maragorik, PNG; Mt
Kasi, Fiji; Peak Hill and Dobroyde, eastern Australia), demonstrate that these systems formed at
shallow epithermal levels.

White (1991) categorised high sulphidation systems on the basis of morphology and alteration
mineralogy/zonations to define the type examples as:
* Nansatsu type as high level disseminated deposits,
* El Indio type which display a structural control,
* Temora type as deeper disseminated deposits;

76
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn

and also emphasised the distinction between early stage alteration and later mineralization. Much
of the morphological differences in Whites classification can be accounted for by wheth-er
magmatic fluid flow, and hence alteration-forming reaction, has been controlled by dilational
structures (e.g., El Indio, Lepanto in White 1991) or permeable lithologies (Nansatsu, Temora,
Peak Hill).

High sulphidation systems are categorized as:

* porphyry-related
* lithologically controlled
* structurally controlled

Lithology and structure are end-members of a continuum of fluid control which in many high
sulphidation systems displays a combination or variation between these elements. The distinc-
tion between epithermal and mesothermal systems is commonly transitional and refers more to
distal of proximal relationships to porphyry source rocks than to crustal levels of formation.

It appears that some high sulphidation systems formed in distal settings to magmatic-source
rocks may undergo sufficient mixing with groundwaters to evolve into a low sulphidation style
of fluid. Exhalative high sulphidation systems are also distinguished. Thus other high sulphida-
tion systems are categorized as:
* composite
* hybrid
* exhalative.

Figure 6.1 illustrates the main styles of high sulphidation systems showing also a relationship to
depth of proximity to the magmatic source.

c) Active Analogues

Fluids enriched in volatile components (H2O, CO2, SO2, Cl, F, B), which are channelled up ma-
jor crustal faults can migrate directly from a degassing magma to the surface and vent as solfa-
taras or fumaroles (Fig. 6.1). Disproportionation of these gases within the fault zones produces
very hot and highly acidic fluids. Fumaroles associated with the White Island andesite volcano
in New Zealand vent gases and acidic fluids at temperatures of up to 600°C, and actively pre-
cipitate native sulphur deposits. The magmatic fluid discharge from the 1988 eruption at White
Island, New Zealand has been calculated at 110 tons/year copper and >36 kg/year gold (le
Cloarec et al., 1992). Thousands of ppm copper and arsenic, and anomalous gold occur within
the deposits derived from the active Surimeat solfatara on the island of Vanu Lava, Vanuatu
(Leach, unpubl. data).

At Biliran Island, Philippines, magmatic volatiles vent to the surface at the Vulcan solfatara in
the form of superheated steam and magmatic gases, and produce liquid sulphur flows up to 1-2
km long (Mitchell and Leach, 1991). This magmatic, gas-dominated fluid has been emplaced
within a pre-existing deep circulating (low sulphidation) geothermal system, which has incor-
porated some of the magmatic volatiles (e.g., Fl" is an order of magnitude higher than other
Philippine geothermal systems). Feeders to the magmatic solfatara were intersected by drilling
at depths of 1 km, and encountered fluids at >310°C and pH <2.

77
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

d) Alteration Assemblages

The range in alteration mineralogy encountered in high sulphidation systems is illustrated in

Figure 6.2. In porphyry environments, high sulphidation alteration zonations are produced by
the progressive acidification, through gradual disproportionation, of reactive magmatic vola-
tiles (mainly SO2 and Cl) as they evolve from a high level intrusion. At shallower levels altera-
tion zonations in high sulphidation systems are formed in response to neutralization and cool-
ing of this hot acidic magmatic-derived fluid through either:
* wall rock reaction, and/or
* mixing with neutral, connate, meteoric or hydrothermal fluids residing in the host
rocks.

The progressive neutralization and cooling of the hot acid fluid results in the formation of
zoned alteration characteristic to high sulphidation systems. Although present in both structur-
ally and lithologically controlled high sulphidation systems, the zonation may be complicated
by overprinting alteration. Four main alterations groupings are classified within the zoned alter-
ation grading outward from the fluid plumbing system, commonly either a permeable horizon
or dilational structure. These are:

i) Vughy or residual silica results of intense host rock leaching by hot fluids at a very low pH
(<2). Only silica (usually quartz, but at shallow levels may be cristobalite, tridymite and/or
opaline silica) and locally some rutile remain under these conditions. While the vughy texture
which is indicative of the leaching generally predominates, silica deposition is locally recog-
nised e.g., Peak Hill, Eastern Australia.

ii) Silica-alunite alteration which rims the silica core contains alunite group minerals (with
zunyite-andalusite ± corundum at near porphyry levels), formed under a slightly higher (2-3)
fluid pH range, as the fluid becomes neutralized in response to wall rock reaction and/or fluid
mixing. The silica and silica-alunite alteration zones comprise what is generally termed ad-
vanced argillic alteration.

iii) Argillic alteration characterised by of kaolin group (pyrophyllite, dickite, kaolinite) miner-
als which are indicative of a formation at a pH of around 4 in turn rims the alunite. These
grade outwards to illite group (sericite, illite, illite-smectite, smectite) minerals (pH around
5), as the fluid becomes progressively more neutralized. The mineral assemblages formed in
each zone are dependent upon temperature and pH of the upwelling acid fluid, the composition
of the host rock, and the physicochemical conditions residing in the host rock.

iv) Propylitic alteration which forms peripheral to the acid alteration assemblages is charac-
terised by sub-propylitic chlorite-carbonate assemblages or propylitic epidote/actinolite-
albite-chlorite-carbonate assemblages. In some systems this mineralogy overprints earlier
formed calc-silicate mineralogy.

The alteration zonations formed in response to upwelling acidic magmatic-derived fluids are
distinct from those formed by descending acid sulphate waters. The latter produces alteration
zones which are indicative of a change from cool and acid to hot and neutral conditions (Fig.
6.2). Failure, to distinguish between these two alteration patterns could result in the misidenti-
fication of peripheral portions of high sulphidation alteration as caps to adularia-sericite epi-
thermal vein systems.

78
 High
Sulphidation Systems
Alteration Mineralogy

1- Porphyry high sulphidation systems

2. Structural high sulphidation systems

a. Silica core

b. Peripheral zones

3. Lithological high sulphidation systems

4. Descending cool acid sulphate fluids

 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

e) Two stage alteration and mineralization model

High sulphidation systems are characterised by two separate stages of alteration and minerali-
zation (Fig. 6.3). Initial leaching of the country rock as the fluid becomes progressively cooled
and neutralized, is interpreted to have been caused by a volatile-rich acidic fluid (White, 1991).
This results in the formation of alteration which grades outward as concentric zones away from
the fluid plumbing system, described above as: a vughy or residual silica core, silica-alunite, to
argillic, and peripheral propylitic alteration.

This initial phase of volatile leaching is commonly followed by a second phase event which may
be polyphasal and comprises an early stage advanced argillic to argillic alteration and as-sociated
deposition of gangue phases; mainly quartz, alunite and/or barite, followed by deposi-tion of
sulphides in breccia zones, cavities and vughs. This event has been interpreted by White (1991)
to be an liquid-rich acidic fluid but a less reactive event. This may be partly because the host
rocks are already altered. Native sulphur infills open cavities and fractures in some sys-tems, and
is the last phase to be deposited, as sulphur displays a low melting point (113°C).

This liquid-rich phase utilises the same plumbing systems as the earlier volatile-rich phase and
focuses mineralized fluids into the vughy or residual silica at the core of the zoned altera-tion.
Competent residual silica and silica-alunite rocks brecciate well and so host mineraliza-tion
(e.g., Nena, PNG, Bainbridge et al., 1994). Continuing deformation of dilational structures
channels the liquid-phase fluids may enhance the formation of fluidised hydrothermai injection
breccias (Section 3.ix.d 1). Gold-copper grades may be proportional to the quantity of breccia
matrix in styles of breccias (e.g., Mt Kasi, Fiji, Corbett and Taylor, 1994). The enclosing in-
competent clay alteration generally displays more plastic deformation, does not fracture, and so
is commonly not mineralized. A skin of barren silica-alunite alteration may rim the mineral-ized
silica core and mask mineralization, especially if this material is relatively hard (e.g., Ne-na,
PNG, Fig. 6.22). In some systems the clay alteration has a damning effect and so the inter-face
between the competent and incompetent rocks may represent a locus for higher metal grades
(e.g., Binebase, Sangihe Is, Indonesia; Corbett, unpubl. data). Elsewhere, (e.g., Wafi, PNG)
peripheral clay zones formed during the liquid event and became sites of maximum mix-ing and
local high gold grades. Overprinting zoned alteration is recognised in many systems as a result
of this two phase alteration (Fig. 6.3).

The sequence of early leaching, zoned alteration and silicification followed by fracturing,
brecciation and mineralization suggests that in most high sulphidation systems there has been
selective partitioning of metals into the liquid phase during melt crystallization (contrary to the
findings of Heinrich et al., 1992). However, significant gold ± copper mineralization is locally
encountered in the clay zones formed during the initial alteration event (e.g., Wafi Zone A,
Lepanto stratiform ore). This is interpreted to indicate that under certain conditions (i.e., mul-
tiple influxes of hot acid fluids), some metals may initially partition into the volatile phase,
since the composition of that phase is interpreted to change progressively with time (Candela
and Piccoli, 1995).

f) Mineralization

The locally polyphasal sulphide mineralization infills open space and forms breccia matrix
comprises early deposition of iron sulphide, followed by copper sulphide phases. Such a se-
quence of initial iron sulphides and later base metal sulphides is also characteristic of low sul-
phidation intrusive-related systems, especially in the deeper level quartz-sulphide vein sys-
tems. This may reflect sequences of early iron and later copper fractionation from the melt

79
 Fig. 6.3

Zonations in Metals and Copper Sulphides in High Sulphidation Systems

 Exploration Workshop 'Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

during crystallization.

The iron sulphide phase is composed predominantly of pyrite at all depths, whereas marcasite
and melnicovite-pyrite are common only at shallow levels. Arsenean pyrite is encountered only
in settings of marginal argillic and propylitic alteration. In some systems the pyrite phase is a l-
so polyphasal with early coarse pyrite, overprinted by later fine pyrite-quartz, commonly form-
ing as banded massive sulphide veins or the matrix to breccias.

Copper deposition is typically the last event, and occurs infilling open vughs, and within frac-
tures and breccia zones, generally overgrowing earlier pyrite and gangue phases. High sul-
phidation systems exhibit zonations in metals and sulphide phases, both laterally and vertically
from deep levels proximal to intrusive sources to higher crustal levels, and from silica cores to
peripheral argillic alteration (Fig. 6.4). The Cu:Au ratios decrease from deeper porphyry levels
to higher epithermal levels. At intermediate depths the systems are arsenic-rich, and at very
shallow near surface levels, high sulphidation systems commonly exhibit enrichment in telluri-
um, antimony and locally mercury. The central silicic zones are copper-arsenic-rich, whereas the
marginal argillic-propylitic zones are dominated by lead-zinc mineralization.

There is a corresponding zonation in sulphide phases which reflect the metal zonations (Fig.
6.4). Hypogene covellite is commonly the main copper sulphide phase at levels proximal to the
intrusive source (e.g., porphyry zone at Wafi), whereas enargite/luzonite are encountered at more
distal environments (e.g., Nena, PNG; Lepanto, Philippines). A transition from enargite to its
lower temperature polymorph luzonite occurs in cool outflow zones (e.g., Nena, Fig.
6.23). Antimony, tellurium, vanadium and mercury substitute for copper and arsenic at shallow
epithermal levels to form phases such as stibioluzonite, goldfieldite, sulvanite and schwazite
respectively. The formation of antimony, vanadium, tellurium and mercury phases at epither-
mal levels in high sulphidation systems is comparable to the abundance of similar metal phases
at shallow levels in many intrusive-related low sulphidation systems. This may reflect the ex-
tensive dilution of the magmatic fluid at distal settings to the intrusive source, by mixing with
groundwaters (e.g., Mt Kasi, Fiji; Leach, unpubl. data).

The copper sulphide phases become progressively more iron rich moving from the central
silicic zones to marginal argillic zones (Fig. 6.4), e.g.,
covellite----- > chalcopyrite at deep levels, and
enargite/luzonite ------- > tennantite ------ > chalcopyrite at shallow levels.

The base metal phases galena and sphalerite are encountered in peripheral zones distal to the
silicic cores.

High sulphidation systems in the southwest Pacific are commonly silver -poor. Gold is typi-
cally of a very high fineness (> 900), and occurs at deeper levels as submicroscopic inclusions
in sulphides, or in the lattice of sulphides, and as free native gold or Au-tellurides at shallower
levels. Gold is commonly associated with the copper phases, however in some systems signifi-
cant gold is also deposited in later pyrite phases (e.g., Zone A at Wafi, PNG), especially those
which immediately predate copper mineralization. High sulphidation systems formed at higher
crustal levels (e.g., Mt Kasi, Fiji), and the upper oxidised portions of most high sulphidation
systems (e.g., Nena, PNG) typically display better metallurgical characteristics.

80
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

ii) High Sulphidation Systems Formed as Shoulders to Porphyry Intrusions

a) Characteristics

Porphyry-related high sulphidation systems may be encountered immediately adjacent to the

intrusive source from which the acidic fluids have been derived (e.g., Horse-Ivaal, Frieda Riv-er,
PNG; Lookout Rocks, New Zealand; Cabang Kiri, Indonesia). As fluids commonly exploit
fracture permeability above the margins, rather than the centre of intrusives, the term shoulder is
preferred to caps, as a description of the setting for porphyry-related high sulphidation alter-ation.
While these roughly correspond to the porphyry copper lithocaps defined by Sillitoe (1995b),
greater emphasis is placed here on the structural control as ledges (also utilised by Sil-litoe,
1995b) on intrusion margins than as overlying caps. In these systems magmatic volatiles may be
released into the same bounding structures initially utilised by the source intrusive dur-ing its
emplacement. Through progressive disproportionation of reactive magmatic volatiles such as SO2
and Cl, the fluids become gradually more acidic and cooler, moving away from the source
intrusive, both vertically and horizontally. The formation of advanced argillic zones marginal to
high level intrusives is interpreted to take place during early stages of development of the
porphyry-related hydrothermal system. In active porphyry systems in the Philippines, the
production of hot acidic fluids occur after zoned propylitic-potassic alteration (e.g., Alto Peak),
but pre-dates the collapse of meteoric waters to form overprinting phyllic alteration.

The gradational increase in fluid acidity results in a zonation of alteration mineralogy away
from the porphyry as:
* potassic grading to phyllic alteration proximal to the porphyry,
* advanced argillic alteration dominated initially by andalusite and then pyrophyllite +
diaspore,
* alunite in silicic zones in distal settings.
The silicified advanced argillic zones commonly form ridges (ledges) distal from the intrusive,
whereas the potassic-phyllic zones occur in more deeply eroded valleys proximal to the intru-
sion. Thus, these systems commonly form rugged topography in regions of high rainfall and
erosion (e.g., Lookout Rocks, New Zealand; Horse-Ivaal, Frieda River, PNG). Fractured intru-
sive margins typically act as fluid conduits in dilatant fractures adjacent to the intrusive (e.g.,
Horse-Ivaal, Frieda River, PNG; Batu Hijau, Indonesia, Meldrum et al., 1994; Lookout Rocks,
New Zealand, this manual). The term ledges is commonly used for silicified zones which dis-play
tabular morphologies, commonly by the exploitation of a dilatant structure or permeable lithology
(Sillitoe, 1995b). Silica tends to form as locally brecciated pervasive silicification and lacks the
vughy character of residual silica in ore-related high sulphidation systems. Corundum is locally
encountered in high temperature zones immediately adjacent to some intrusions. Very coarse
grained muscovite may be associated with the advanced argillic alteration miner-alogy, and the
mutual intergrowth with alunite is diagnostic of high sulphidation systems at porphyry depths
(e.g., in the diatreme at Wafi). Some high sulphidation systems proximal to source intrusives are
enriched in halogens derived from the magmatic volatiles. This is pro-nounced in those systems
sourced from felsic intrusives, and may be exhibited by the for-mation of mineral phases such as:
tourmaline, topaz, dumortierite, apatite, and zunyite.

Porphyry-related advanced argillic shoulders are typically barren of significant mineralization,

possibly due to the inability of the magmatic volatiles to transport metals (see above). Through
careful mapping of alteration zonations it can be determined that these types of high sulphida-
tion systems are genetically associated with the immediately adjacent intrusives. However, as
will be outlined in Sections 6.iii and 6.iv, in some cases high sulphidation-style

81
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure. Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

alteration overprints earlier porphyry-related alteration, and therefore the hot acidic fluids are
interpreted to be derived from a deeper source than the host porphyry stock. Significant copper-
gold mineralization commonly occurs in systems where earlier porphyry-copper intrusives are
overprinted by later high sulphidation fluids (e.g., FSE-Lepanto, Philippines; Wafi-Rafferty's,
PNG).

b) Examples

i) Horse-Ivaal, Frieda River Copper, Papua New Guinea

The Horse-Ivaal porphyry copper deposit, within the Frieda River porphyry complex of
northwest PNG, is described in detail by Britten (1981) and summarised by Asami and Britten
(1980) and Hall et al., (1990). The Horse-Ivaal deposit is centred around the fine grained Horse
microdiorite which has been emplaced into older diorite porphyry. Copper mineraliza-tion,
mainly as chalcopyrite, is associated with late sericite-chlorite + anhydrite alteration and
veining, which postdates quartz - anhydrite stockwork veins and alteration which is zoned from
potassic (biotite ± K feldspar) to propylitic.

Two zonations in alteration mineralogy are evident at Horse-Ivaal (Figs. 6.5, 6.6):

1. There is a gradual change from a potassic zone characterised by biotite, which grades
laterally (in the deeper parts of the system), to a propylitic alteration containing albite-epidote.
These alteration assemblages were formed during the early stages of development of the
hydrothermal system and are indicative of a progressive decrease in temperatures (possibly
conductively) moving away from the main heat source.

2. There is a progressive change south and west from;

* the central biotite zone,
* to a transitional zone characterised mainly by a K-feldspar - chlorite overprint on
biotite,
* grading outward either to a sericite and/or chlorite zone,
* to zoned peripheral advanced argillic alteration.
The change from the sericite to the advanced argillic alteration is characterised by the initial
appearance of andalusite, followed by a gradual change from sericite to pyrophyllite ± diaspore
and increased silicification. The pyrophyllite-diaspore zone then grades outward to an alunite
zone. Figure 6.2 illustrates that this zonation is indicative of an initial progressive decrease in
fluid pH, interpreted to be due to the gradual disproportionation of magmatic volatiles, which
facilitates the formation of acidic fluids within the shattered carapace to the intrusive stock.

The shoulder of advanced argillic alteration formed peripheral to the Horse-Ivaal porphyry
deposit, is typical of many other similar zones of high sulphidation alteration in the southwest
Pacific, in which the fluid chemistry is dominated by sulphur in the form of SO2. The unminer-
alized nature of the Horse-Ivaal high sulphidation alteration is probably due to the inability of
the magmatic volatiles to transport significant metals. Local copper mineralization hosted in
advanced argillic alteration at Ivaal is associated with later cross cutting chalcopyrite veinlets.
However, in some cases the high sulphidation systems in close proximity to porphyry deposits
exhibit high grade copper ± gold mineralization (e.g., Wafi River, PNG; FSE -Lepanto, Dizon,
Philippines; Butte, USA). As illustrated in the examples below, careful mapping of the altera-
tion zonation is required in order to distinguish barren from potentially mineralized high sul-
phidation systems formed in association with porphyry systems.

82
Alteration Zones in the Horse-lvaal Porphyry Copper System, Frieda River, PNG.

Fig. 6.5

Alteration Zonational Along NE Section Line A-A' Through North-Western Portion of

the Horse-lvaal Porphyry Copper System, Freida River, PNG.

Fig. 6.6
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

ii) Lookout Rocks. New Zealand

A jog in the Hauraki Graben Fault which separates the Hauraki Graben from the Coromandel
Peninsula hosts; the Ohio Creek copper-gold porphyry, the Lookout Rooks high sulphidation
shoulder, and the Thames 2 million ounce mesothermal vein system (Figs. 7.44, 7.45). Dextral
rotation on the graben structures, associated with plate rotation, has produced a series of dila-
tional fractures which host ledges of high sulphidation alteration formed as shoulders to the
Ohio Creek porphyry (Fig. 6.7) and the auriferous mesothermal veins in a more distal setting
(Fig. 7.45).

Scout drilling in the 1970s defined copper-gold-molybdenum mineralization associated with the
Ohio Creek porphyry (Merchant, 1986). The copper mineralization occurs as narrow quartz
stockwork veins hosted typically near the intrusion contacts in a quartz diorite porphyry stock
which has been emplaced into andesite volcanics. The mineralization does not extend into
Lookout Rocks to the south east (Fig. 6.7). Mesothermal auriferous quartz-sulphide veins cut the
Ohio Creek porphyry (e.g., Kaiser Reef) and were the target of small scale mining last cen-tury.

Immediately southwest of the Ohio Creek porphyry, a number of discrete dipping ledges of
advanced argillic alteration form an arcuate series of silicified ridges extending over an eleva-
tion of several hundred metres, and for a strike distance of over 1000 m (Figs. 7.45, 6.7). There
is a progressive zonation in alteration mineralogy from the NW to SE (Figs. 6.7, 6.8), and
grading from depth within the intrusion, to shallow levels along the margins of the intrusion and
extending into the volcanics, as:
* propylitic - epidote-chlorite,
* intermediate argillic - chlorite-sericite,
* phyllic - sericite-quartz,
* advanced argillic - pyrophyllite-diaspore-quartz
* silicified ledges with progressively more abundant alunite (in alunite-pyrophyllite)
which form the Lookout Rocks pinnacles,
* argillic - dickite-kaolinite and illitic clays,
* low grade propylitic - chlorite-carbonate ± illitic clays.
Andalusite, trace tourmaline and secondary apatite occur in association with sericite, and lesser
pyrophyllite. Continuing to the south and east (Figs. 6.7, 6.8).

The zonations in alteration mineralogy are indicative of a gradual decrease in fluid pH from
deeper porphyry levels to shallow levels at the Lookout Rocks. Cooling and neutralization of
this hot acidic fluid continued further to the south and east in the argillic and propylitic altera-
tion. The presence of andalusite, tourmaline and apatite demonstrate that the alteration at Look-
out Rocks took place at high temperatures from fluids containing a significant magmatic com-
ponent. It is interpreted therefore that the changes in fluid pH resulted from the progressive
disproportionation of reactive magmatic volatiles emanating from the magmatic source. These
volatiles have been channelled into the dilatant structures to form the ledges by reaction with
ground waters. The occurrence of mesothermal quartz-sulphide veins and of late stage andesite
dykes which cut the Ohio Creek porphyry, suggest that either a later intrusion may have been
emplaced at depth, or that the parent melt to the porphyry exsolved fluids and metals from depth
(Figs. 6.7, 6.8). Gold mineralization in the mesothermal veins is discussed in Section 7.iv.d.2.i

83
Fig. 6.7

Fig. 6.8
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

iii) Vuda, Fiji

An area of high sulphidation alteration in the Vuda Valley, eastern Viti Levu, Fiji (Fig. 6.9), has
been subject to exploration for porphyry copper and later epithermal gold targets. Minor gold
production is recorded from several small, mainly pre-World War II mines located out-side the
high sulphidation alteration. Some of the early exploration (Corbett unpubl. data; Austpac Gold
Prospectus, 1985), has been reinterpreted in the light of new models. Late Mio-cene shoshonitic
lavas, autoclastic breccias dominate as host rocks and recent gravity data is consistent with the
extension of intrusive rocks which outcrop in the region to below the alter-ation zone (Colley and
Flint, 1995). The alteration is zoned about feeder structures which occur as arcuate siliceous
ridges and may represent caldera ring fractures. Alteration grades laterally through silica-alunite-
kaolinite, to pyrophyllite-diaspore and clay dominant assemblages (Figs. 6.9, 6.10). The silica-
alunite zone contains fine quartz crystals and weathers to a characteristic crunchy soil under foot.
Diaspore is also coarse grained and euhedral. Peripheral clay alteration exhibits rings derived
from soaking into the rock adjacent to fractures. Lower temperature alter-ation including
chalcedonic silica to the north, and the overall shape of the system, are indica-tive of possibly
tilting of the system, to expose deeper alteration assemblages in the southern portion (Fig. 6.9).

Alteration studies in association with an extensive programme of trench sampling and diamond
drilling in 1985/86 by the Vuda Joint Venture partners, demonstrated that the alteration formed at
too high a temperature to be regarded as epithermal (in the then terminology) and is essen-tially
barren. Most gold mineralization occurs outside the high sulphidation alteration, typically with
adularia on fractures, and within vughs in relatively fresh rock. Minor alteration-hosted gold
may have formed by the overprinting of clay alteration on existing mineralization. The Natalau
mine was the main producer with a production of 880 oz Au to 1954 (Colley and Flint, 1995).
Here gold is associated with pyrite and base metal sulphides (S. Henderson, pers commun.,
1993) on fractures at the margin of a competent mafic dyke. Lawrence (1984) de-scribes the
fineness of the Natalau gold as 850, typical of formation in a low sulphidation mesothermal
environment, rather than an epithermal setting (see Fig. 4.8). Drilling intersected alteration
typical of porphyry environments as phyllic alteration characterised by a sericite-anhydrite-
quartz assemblage, best developed in structures reactivated as clay matrix breccias, and
propylitic alteration in the footwall of the feeder structures.

It is proposed (Fig. 6.10) that the high sulphidation alteration formed within feeder structures
above (subjacent to) a porphyry, evident on the gravity data (Colley and Flint, 1995). Gold
mineralization is interpreted to be of the mesothermal quartz-sulphide-style, typical of that
which forms peripheral to many Pacific rim porphyry intrusives (Section 7.ii). The presence of
gold mineralization in association with adularia on fractures and within vughs, especially on the
periphery of the system, led to the interpretation of an epithermal origin. However, potassi-um-
rich shoshonitic host rocks readily deposit secondary K-feldspar, and so the extensive wallrock
adularia alteration need not be indicative of gold deposition by boiling in an epither-mal
environment.

iv) Cabang Kirif Indonesia

Copper mineralization at Cabang Kiri, North Sulawesi, is hosted in quartz stockwork veins
which extend from cylindrical quartz diorite porphyry stocks into surrounding metavolcanics
(Carlile and Kirkegaard, 1985). Advanced argillic alteration occurs as pervasive alteration
within the porphyry intrusives, and structurally controlled in the andesitic host rocks. Altera-
tion is described by Lowder and Dow (1978) as grading outwards through the following

84
Fig. 6.9

Fig. 6.10
 1
Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization ' Corbett G J & Leach T M, 8/96 Edn.

assemblages:
* andalusite-corundum-muscovite,
* andalusite-pyrophyllite/muscovite,
* andalusite-pyrophyllite-diaspore,
* diaspore-pyrophyllite,
* diaspore-pyrophyllite-alunite,
* alunite-kaolinite-pyrophyllite,
* silica-alunite-kaolinite.
This zonation in high sulphidation-style alteration is indicative of progressive cooling and de-
crease in fluid pH away from the porphyry intrusions (Fig. 6.2). In a manner similar to Horse
Ivaal and Ohio Creek-Lookout Rocks, significant copper mineralization at Cabang Kiri occurs
in andalusite-pyrophyllite zones, but does not extend out into the alunite-bearing assemblages
(Lowder and Dow, 1978).

Hi) Lithologically Controlled High Sulphidation Gold-Copper Systems

a) Characteristics

The controls to fluid flow in most high sulphidation systems are governed by a combination of
lithological and structural controls and so any classification focuses upon end members only. It is
common for fluids to exhibit a structural control in venting from the source intrusion along
dilational structures and then, upon contact with a permeable lithology, migrate along that unit
and so display a lithological control. Lithologically controlled high sulphidation alteration form
as magmatic fluids are focused along permeable host rocks such as coarse grained pyroclastics
within a less permeable sequence (e.g., Nansatsu Deposits in Japan, below; Mt. MacKenzie and
Clive Creek in eastern Australia; Leach, unpubl. data) or permeable sediments and diatreme
breccias (e.g., Wafi in PNG, below). Although the fractured diatreme margin provides a per-
meable host rock at Lepanto, Philippines (Fig. 6.25) fluid is focused into this setting by a dila-
tional structural environment (Fig. 6.24) and so this classed a predominantly structural control.
Disseminated ores are common in lithologically controlled high sulphidation alteration.

In deeper level systems the diffusion and slow neutralization and cooling of upwelling hot acidic
fluids may facilitate the formation of relatively broad alteration zones (e.g., Wafi, PNG).
Quenching in higher level systems results in the formation of sharp alteration boundaries (e.g.,
Nansatsu deposits). Although most mineralization is disseminated filling open space or as frac-
ture/breccia ores within the silica core or silica-alunite alteration zones, some broad alteration
systems host gold in the halo of clay alteration (e.g., Wafi, PNG).

b) Examples

i) Wafi River Prospect, PNG

The Wafi River Prospect is predominantly a lithologically controlled style of high sulphidation
gold system which is hosted in low grade Mesozoic Owen Stanley metasediments (siltstones,
litharenites and conglomerates) and a diatreme breccia. The following discussion on the Wafi
River high sulphidation system is taken from Leach and Erceg (1990), Erceg et al., (1991), and
CRA (1994).

The Wafi Transfer Structure is inferred to have focused overprinting emplacement into the
metasedimentary host rocks multiple phases of microdiorite and quartz diorite porphyry

85
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

intrusions and a diatreme breccia complex (Fig. 6.11; Corbett, 1994). The magnitude of this
structure is evident in the manner in which it separates the western and eastern segments of the
New Guinea Orogen and offsets the Markham Fault, a plate boundary to the north of Wafi
(Corbett, 1994). Although NW trending accretionary structures display a clockwise rotation by
the sinistral displacement of the NNE trending transfer structures, other senses of movement are
possible during the life of the system.

Surface alteration comprises a 4 km2 ares of concentrically zoned advanced argillic and argillic
alteration (Fig. 6.12). It is interpreted that the primary permeability in the sediments (porous
coarse grained units, bedding planes and metamorphic quartz veining), enhanced by multiple
fracturing events, and in the diatreme complex, has facilitated the dissemination of hot acid
fluids over this large area. Alteration ranges outwards dominated by the assemblages:
* residual silica,
* quartz-alunite,
* alunite-pyrophyllite or alunite-dickite,
* pyrophyllite or dickite,
* dickite/kaolinite—illitic clay
zones, *interlayered clay,
* chlorite.
This zonation is indicative of a progressive cooling and neutralization of a hot acidic fluid which
was sourced from the east and radiated outward (Leach and Erceg, 1990), to provide the
elongate alteration (Fig. 6.12). Zones of significant gold mineralization are encountered around
the margins of the diatreme complex (Zones A, B, and D, Fig. 6.12), in structurally controlled
breccias (Zone C) which crosscut the diatreme complex, and in distal settings (Malaria and
Hesson Creek) aligned along the NE-trending structures. Gold mineralization occurs in two
settings and is interpreted to be related to two separate events:

i) An initial phase of disseminated auriferous pyrite mineralization within alunite-dickite and

clay zones which is inferred to be related to the initial formation of the zoned advanced argil-lic
alteration, and associated with the influx of hot acidic fluids along the margin of the miner-
alized porphyry.

ii) A later phase hosted in pyritic fractures and breccias which crosscut the zoned alteration.

Gold mineralization in the Pueblo Viejo high sulphidation system, Dominican Republic, is
similarly attributed to two pulses of mineralized magmatic fluids: one associated with dissem-
inated pyrite in clay-alunite alteration zones at depth, and the other hosted in pyritic veins at
shallower levels (Muntean et al., 1990). At Lepanto, copper-gold mineralization occurs asso-
ciated with early clay alteration in stratiform deposits, and in later fractured silicified zones (see
Fig. 6.25).

Gold in Zone A at Wafi (>15 Mt at 2.6 g/t Au) is refractory and generally submicroscopic,
although a few minute (1-3 micron) inclusions have been observed in pyrite both disseminat-ed
in the altered sediments, and infilling fractured and brecciated metamorphic quartz veining.
Copper mineralization in Zone A occurs in trace amounts as enargite and luzonite in the quartz-
alunite-dickite zones, and as tennantite, with base metal sulphides, in the peripheral argillic
zones. Fluid inclusion data on sphalerite associated with the acidic alteration indicates that
mineralization took place at cool (200-220°C) epithermal levels.

A blind mineralized porphyry stock was encountered at Wafi 800 m NE of Zone A, beneath a
leached cap in the region of the inferred upflow of acidic fluids (Erceg et al., 1991). Copper
mineralization occurs predominantly as hypogene covellite, in places intergrown with

86
 Fig. 6.11

Fig. 6.12
Fig. 6.13

Fiq. 6.14
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

chalcopyrite, bornite and pyrite. It appears (Erceg et al., 1991) that the hot acidic fluids were
sourced from depth and migrated along the margin of the high grade mineralized quartz diorite
stock (Rafferty's Porphyry). The initial drill intercept in the porphyry copper intrusive, drill
hole WR 95, yielded published results of 263 m at 1.86 percent Cu and 0.27 g/t Au.

Two main episodes of hydrothermal activity are therefore recognised within the Wafi prospect
area (Fig. 6.14):

i) A porphyry copper event which involved emplacement of the Wafi porphyry into the Wafi
Transfer Structure.

ii) A high sulphidation event followed uplift of the porphyry resulting in widespread over-
printing of the earlier porphyry system. A similar acidic hypogene phase at Butte, Montana
(Brimwall and Ghiorso, 1983) and at El Salvador (Gustafson and Hunt, 1975) has been inter-
preted by these authors to have remobilized pre-existing copper protore to form a secondary
covellite-chalcocite ore.

ii) Nansatsu Deposits, Japan

The Nansatsu deposits are located in southern Kyushu, Japan, This text is derived from reviews
of the Nansatsu Deposits (Hedenquist et al., 1988, 1994; Matsuhisa et al., 1990; White 1991;
Izawa and Cunningham, 1989) and personal observations (Corbett, unpubl. report, 1987). The
deposits are characteristically small, mushroom-shaped bodies, with the three current produc-ers,
Kasuga (0.15 M oz Au), Iwato (0.21 M oz Au) and Akeshi (0.22 M oz Au). All exhibit low gold
grades in the order of 3-4 g/t Au, but locally contain higher gold grades within feeder structures
such as Kasuga (Hedenquist et al., 1994) or breccias (Izawa and Cunningham, 1989). The silica-
rich ores are used as flux in the copper smelters, from which the gold is ex-tracted.

Ages of alteration similar to the Upper Miocene-Pliocene host volcanic sequence, rapid changes
of the marginal alteration, and presence of interpreted explosion breccias (Izawa and
Cunningham, 1989) all suggest that the Nansatsu deposits formed at relatively high crustal
levels. Hot acid magmatic-sourced fluids migrated from feeder structures into more permeable
pyroclastic units in the predominantly lava sequence of volcanics, to form tabular or mush-
room-shaped silicified bodies (Fig. 6.15). Eruption breccias (Izawa and Cunningham, 1989)
provide additional permeability. Cooling and neutralization of those fluids by rock reaction is
reflected by a characteristic zoned alteration pattern (Fig. 6.15) which grades from: the core of
residual silica through alunite-kaolinite, to the rim of illite and illite-smectite clays with
commonly sharp contacts resulting from pH changes (Hedenquist, pers. comm.). Gold occurs
within the residual silica in association with pyrite, enargite (luzonite), covellite, native sul-phur
and later iron oxides and displays higher grades in the eruption breccias (Izawa and Cun-
ningham, 1989). Fluid inclusion and clay alteration studies suggest mineralization tempera-tures
consistent with the epithermal environment of 170-210°C (Hedenquist et al., 1994; Izawa and
Cunningham, 1989), varying to locally higher temperatures (250-300°C) within deeper levels at
Kasuga (Hedenquist et al., 1994).

iii) Miwah, Indonesia

The Miwah high sulphidation system is described by Williamson and Fleming (1995) and
Leach (unpubl. report, 1995) from which this discussion is taken. Although Miwah displays
characteristics similar to both the Lepanto and Wafi high sulphidation systems, it is classified
as exhibiting a predominantly lithological control. Miwah is located in northern Sumatra,

87
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

Indonesia, in a region of dextral strike-slip faulting related to the Sumatra Fault System (Fig.
1.2).

Alteration and gold mineralization are hosted in a sequence of andesitic to dacitic lavas and tuffs
of the Pliocene Leuping Volcanics. These volcanics are aligned ENE along the Miwah lineament,
and are mirrored by a similar lineation of recent active volcanoes to the north. Dila-tion on ENE
structures is inferred from rotation of the Sumatra Fault system. The Leuping Vo l-canics have
been intruded by porphyritic andesite to rhyodacite dykes and domes (Fig. 6.17) which are dated
by K-Ar at 2.9 m.y. The dykes and domes contain a wide variety of xenolith clasts which range
from andesite and diorite porphyry, magnetite-rich skarns and calcsilicate rocks. In the south and
west regions of the prospect, the volcanics are intruded by a diatreme breccia complex which
contains local dacitic material and quartz-veined andesite clasts. Some of the quartz in the veins
contain anhydrite and halite daughter crystals associated with liquid-and vapour-dominated fluid
inclusions, indicative of formation in an environment proximal to a high level intrusion.

The volcanics, domes, dykes and diatremes have been overprinted by extensive advanced argil-
lic - argillic alteration which is zoned grading outwards as assemblages dominated by:
* vughy to dense quartz-rutile-pyrite,
* quartz-alunite,
* quartz-kaolinite,
* illite-smectite,
* chlorite/chlorite-smectite.
This alteration overprints earlier propylitic, and locally phyllic, alteration. The silicified quartz
and quartz-alunite zones (Figs. 6.16) occur in:
* Restricted zones within inferred dilational NNW trending structures which parallel
the Rusa Fault and crop out on the eastern margins of the prospect.
* Less dominant NNE trending structures which crop out as thin ridges and parallel the
Camp Fault.
* Broad zones within the diatreme breccias, possibly as a reflection of the high primary
porosity in the breccia matrix.
* Shallow (up to >100 m thick) north to northeast dipping zones, hosted in volcanic s.

The quartz and quartz-alunite have acted as brittle host rocks to subsequent fracturing and
brecciation and associated mineralization which changes from early pyrite-rich quartz veining,
to later veining and breccia zones composed of brassy pyrite, overgrown by copper sulphide
phases. The copper mineral phases are dominated by luzonite at shallow levels to the south, and
enargite at deeper levels to the north. Hypogene covellite has been detected locally at depth,
whereas tennantite occurs in more distal settings to the east. The copper phases are in-tergrown
with quartz and banded chalcedonic quartz, and locally at depth with alunite. Native sulphur
commonly infills open cavities and fractures. The alteration and mineralization indicate
relatively cool conditions during the high sulphidation system.

Although there is a close relationship between gold and copper-arsenic contents, gold miner-
alization is not always associated with enargite/luzonite, and so may have been deposited with
earlier pyrite. In recent drilling, Cu:Au ratios increase with depth and to the north. William-son
and Fleming (1990) suggest that a porphyry intrusion may yet be identified as a source for the
high sulphidation system. Information from the structure, alteration and mineralization indicate
a possible source for hot acidic, mineralized fluids from the north and at depth below the
diatreme breccia, and fluid outflow towards the south.

88
Fig. 6.16

Fig. 6.17
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

iv) Structurally Controlled High Sulphidation Gold-Copper Systems

a) Characteristics

Structurally controlled high sulphidation systems result from a control on magmatic fluid flow
by dilational fault/fracture systems provided, as described in detail below, by: the intersection of
permeable lithologies (e.g., Nena, PNG), margins of diatreme breccia bodies (e.g., Lepanto,
Philippines), en echelon gash veins within structural corridors (e.g., Mt Kasi, Fiji), or combina-
tions of these and other factors.

In these systems central vughy silica and marginal silica-alunite assemblages in cross-section
form bulbous alteration zones surrounded by thin argillic zones which grade out into regional
propylitic alteration. Laterally elongate silica-alunite ridges trend kilometres along controlling
structures (e.g., Nena, PNG). The overprinting relationships of the alteration derived from a
vapour-rich fluid and the subsequent mineralization derived from a liquid-rich fluid may be more
clearly evident in the structurally controlled high sulphidation systems. The utilization of the
same plumbing system focuses mineralized fluid into the core of the zoned alteration where the
competent residual silica readily brecciates in a brittle manner. The surrounding clay altera-tion
is less competent and impermeable and so commonly remains unmineralized and may mask
mineralization (e.g, Nena, PNG). The competency contrast aids brecciation of the brittle rocks.
Breccias categorised as rotational and fluidised breccias (Section 3.ix.d.l) are indicative of fluid
transport in feeder structures and commonly grade to crackle breccias towards the pe-riphery of
the mineralized zones. Gold-copper grades are proportional to the matrix content of breccias and
so tend to fall off moving away from the structurally controlled fluid plumbing systems; that is,
from fluid upflow to outflow zones.

b) Examples

i) Nena, Frieda River Copper, Papua New Guinea

The Nena prospect at Frieda River Copper is an example of a structurally controlled high sul-
phidation system recently described by Bainbridge et al. (1993) and (1994) from which this dis-
cussion is taken. A resource of 45 Mt at 2.7 percent Cu and 0.7 g/t Au has been defined for Nena
to April 1995 (Highlands Gold Limited, press release).

Exploration at Frieda River up to 1983 inferred a porphyry copper resource of 860 Mt at 0.47
percent Cu and 0.31 g/t Au within the Koki and Horse-Ivaal deposits, and 32 Mt at 2.35 per-
cent Cu and 0.58 g/t Au within the Nena high sulphidation deposit, located 6 km northeast of
the porphyry deposits (Hall et al., 1990). An increase in the understanding of high sulphida-
tion gold-copper mineralization and the relationship to buried porphyry copper deposits, in
particular Lepanto, Philippines (Garcia, 1990, 1991) and Wafi, PNG (Leach and Erceg, 1990;
Erceg et al., 1991), facilitated a re-evaluation of the Nena mineralization by Highlands Gold
Limited in the early 1990s.

The Nena Prospect occurs on the margin of the Frieda River porphyry copper intrusive system
which is inferred to have been localised by the NW trending Frieda Fault, formed as a splay
fault from the more regional EW trending Leonard-Schultz Fault (Corbett, 1994; Fig. 6.18). An
inferred dextral rotation has imparted a dilational character to the set.of structures between the
Frieda and Leonard-Schultz faults, and termed the Nena Structural Corridor (Figs. 6.18,
6.19). These structures host a series of silica and silica-alunite ridges which extend for over 10
km from the Horse-Ivaal porphyry copper deposits and include the Nena high sulphidation
system and the Horse-Ivaal barren porphyry shoulder (Figs. 6.19, 6.20). The high sulphidation

89
Fig. 6.18

Fig. 6.19
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

system at Nena occurs as NW oriented concentrically zoned bulbous alteration (Figs. 6.21,
6. 22) which grades outwards as assemblages dominated by:
* vughy (residual) silica,
* quartz-alunite zone (locally sulphur-bearing),
* pyrophyllite-dickite-kaolinite,
* illite-smectite,
* carbonate-gypsum-chlorite.

The alteration is interpreted to have formed from acid leaching by an initial vapour-rich mag-
matic fluid phase (White, 1991) which migrated laterally in the north to south direction along
dilational feeder structures. The gradation from broad central zones of vughy (residual) silica
outward to quartz-alunite alteration is postulated to have formed as a response to the progres-
sive cooling and neutralization of this acidic fluid through rock reaction; whereas peripheral thin
clay zones are suggestive of rapidly changing fluid physico -chemistry upon mixing with
circulating meteoric-dominated fluids. The alteration shows a preference for the permeable vol-
caniclastic units within a sequence interlayered with lavas such that the intersection of the Ne-na
structure with the pyroclastic unit forms the locus of fluid flow.

Copper and gold mineralization are associated with a later, predominantly liquid, magmatical-
ly-derived fluid which has utilised the same feeder structures as the volatile phase, and brecci-
ated the earlier competent vughy silica. Fractures, breccias and open leached vughs have been
sealed by initial multiple phases of pyrite. Copper mineralization occurs as late stage sulphides
deposited in cavities and fractures in the pyrite, in places intergrown, and locally rhythmically
banded with barite. Intense brecciation and local fluidised breccias accompany high grade cop-
per mineralization within the central vughy silica zones; whereas more fracture controlled sul-
phide deposition results in low grade copper-mineralization in the peripheral quartz-alunite
zones.

Initial fluid inclusion studies (Leach, unpubl. reports) on barite associated with copper mineral-
ization indicate that the mineralized fluid was two phase, relatively hot (>300-350°C) and mod-
erately saline (>9-10 wt percent equivalent NaCl). Mineralization developed in response to rapid
cooling upon mixing with low temperature (<150-200°C), diluteo(<l-2 wt percent NaCl)
meteoric waters.

The copper mineral phases exhibit a lateral zonation (Fig. 6.23) from the northwest Mt. Nena
region to the southern regions of the deposit as: hypogene covellite + enargite, through zones of
enargite, and luzonite > enargite, to only luzonite at shallow levels. Covellite is early in the
paragenetic sequence and is locally altered to enargite. Cu:Au ratios in the sulphide zone de-
crease from north to south, and at depth to shallow levels in the southern regions. Luzonite be-
comes progressively Sb- and Te-rich at shallow levels implying solid solution series with
stibioluzonite and goldfieldite.

Supergene leaching has resulted in an oxidised gold zone, overlying a zone of supergene
covellite-chalcocite enrichment. A distinct hypogene gold phase has not yet been identified,
however it is probably associated with the luzonite-goldfieldite series as well as late stage py-
rite. Supergene native gold occurs as very high fineness, minute grains which infill fractures in
the gold-rich oxide zone.

Palaeo fluid flow directions at Nena are given by:

* The gradation in the morphology of the copper-bearing structures from subvertical
feeder zones further north (Fig. 6.21) to concentric outflow pipe-like features (Fig.
6.22).

90
Fig. 6.20

Fig. 6.21
Fig. 6.22

Fig. 6.23
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

* Changes in alteration mineralogy (Fig. 6.23).

* The distribution and characteristics of silica-alunite ridges (Figs. 6.19 and 6.20).
* Zonations in copper phases and Cu:Au ratios.

These directions provide a speculation that the high sulphidation fluids and metals could have
been derived from a high level intrusion at some depth in the vicinity of Mt Nena, and then
flowed laterally to the south along the preferred permeability (Fig. 6.20). This provides for a
lateral fluid flow in the order of at least 1 km.

ii) Lepanto-FSEf Philippines

The Lepanto high sulphidation enargite gold deposit (35 Mt at 3.5 percent Cu, and 3 g/t Au) in
the Philippines is located 200-400 m to the northwest and 400 m above (Fig. 6.24) the high grade
FSE porphyry copper deposit (356 Mt at 0.73 percent Cu, and 1.2 g/t Au).

Structural control for the Lepanto high sulphidation system is provided by the intersection of a
throughgoing dilational structure with fracturing at a flatly dipping diatreme margin (Figs.
6.24, 6.25). The Philippine Fault, formed as a major sinistral transpressional terrain boundary
with a protracted history of activity, breaks up into several splay faults which parallel the
magmatic arc in Northern Luzon (Fig. 2.6; Mitchell and Leach, 1991). NS trending inferred
splays, the Abra and Pseudo Faults (Baker, 1992) constrain NW trending structures (Garcia,
1991) in the mine area (Fig. 6.24). The sinistral rotation presented by Baker (1992) for the
Pseudo Fault coupled with the Abra Fault could dilate the subsidiary Lepanto Fault (Fig. 6.24)
for which Baker (1992) also maps a sinistral rotation. Thus, as shown in Figure 6.24 the
enargite ore is elongate along the dilatant Lepanto Fault.

The pyroclastic and dacite porphyry rocks of Garcia (1991) might also be interpreted to repre-
sent diatreme breccias and endogenous domes respectively (Baker, 1992; Garcia, pers. com-
mun., 1995), and both pre- and post-mineralization diatremes are inferred (Figs 6.24, 6.25). The
outline of the enargite ore in cross section is aligned along the fractured diatreme bounda-ry
(Fig. 6.25), and exhibits an overall elongated plate-like shape.

The FSE porphyry may be localised at the intersection of the Lepanto Fault formed as a splay
from the arc parallel Pseudo Fault (Fig. 6.24), in a manner similar to the inferred localisation
of porphyry mineralization at Frieda, PNG (Fig. 6.18), Chuquicamata, Chile (Boric et al.,
1990), and other Philippine porphyry deposits (Sillitoe and Gappe, 1984).

Porphyry copper-gold mineralization at Guinaoang (Fig 6.24), 6 km to the southeast, is hosted

at depth in a chorite-sericite altered quartz diorite stock, and at shallow levels in advanced ar-
gillic altered volcanics (Sillitoe and Angeles, 1985). A post-mineral diatreme locally cuts al-
tered volcanics at shallow levels.

Garcia and Bongolan (1989), and Garcia (1990, 1991) describe the following events for the
development of the Lepanto and FSE deposits (Figs. 6.24, 6.25, 6.26):

i) Emplacement of the FSE quartz diorite stock (9-11 m.y.) into mid Miocene volcaniclastics
and the development of quartz stockwork veining and zoned alteration. Biotite alteration ex-
tends 100 m from the diorite contact, and grades out to propylitic epidote-calcite-chlorite al-
teration. These have been overprinted by retrograde chlorite—illite/sericite-clay alteration.

ii) A steep hydrothermal breccia pipe transects the FSE diorite, but is truncated at shallow
levels by the Imbanguila Dacite. Alteration in the breccia pipe grades from sericite-tourmaline

91
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

at depth, to anhydrite-alunite (± diaspore, pyrophyllite, zunyite and illite) at shallow levels. This
advanced argillic alteration extends beyond the breccia and forms and extensive cap over the FSE
porphyry. The alteration grades laterally from: quartz-alunite-zunyite in central zones, through
pyrophyllite-diaspore, to margins of illite—chlorite (T. Leach, unpubl. data). The alunite in the
hydrothermal breccia has been dated at 6.9-8.5 m.y. (Hedenquist, unpubl. data).

iii) The funnel shaped Imbanguila dacite porphyry has variably been dated at 5.6-7 m.y. (Gar-cia,
1991) and is associated with a diatreme breccia which mushrooms at shallow levels to the SW
(Fig. 6.18). Advanced argillic alteration and associated gold-copper mineralization prefer-entially
occur at the intersection of faults and fractures (of the Lepanto Fault) and the contact between the
diatreme and the underlying volcaniclastics. Central zones of vughy to massive quartz grade
outward through quartz-alunite-kaolinite, and kaolinite zones to peripheral smec-tite-illite and
chlorite alteration. The dacite porphyry is relatively unaltered except at its mar-gins and in fault
controlled quartz-pyrite-anhydrite zones (Easterlies) grading out to quartz-alunite-kaolinite
alteration.

iv) The Bato Dacite forms porphyry domes and associated tuffaceous diatreme breccias, and
yielded a K-Ar age on biotite of 2.9 m.y. (Sillitoe and Angeles, 1985). Garcia (1991) interprets
these dacite units to be post-mineral. To the west of the mine area, quartz-alunite-kaolinite al-
teration occurs as a 6 km long ridge which defines the lower contact of dacitic units. Alunite in
these ridges (Hedenquist, unpubl. data), as well as sericite in the FSE, have been dated at
between at 1.4-3.3 m.y.

Three main types of mineralization are recognised (Garcia, 1990, 1991) at Lepanto -FSE:

i) Porphyry style chalcopyrite-bornite mineralization hosted within and adjacent to the FSE
quartz diorite is related to the chlorite-illite/sericite-clay overprinting alteration event. Gold
occurs as inclusion in, and overgrowing, copper sulphides. This mineralization is interpreted to
have been either deposited during, and/or remobilized by, the post-FSE hydrothermal breccia
event.

ii) High sulphidation enargite-luzonite mineralization is associated with advanced argillic al-
teration and hosted within central vughy to massive quartz zones: in alunite-anhydrite zones
within the hydrothermal breccia, at the diatreme breccia - metavolcanic contact along the strike
of the Lepanto Fault (branch veins or classical ore), in fault controlled alunite-anhydrite zones
which cut the dacite porphyry (Easterlies), and in flat lying bodies which replace calcar-eous
sediments (stratabound ore). Kaolin clay alteration hosts copper-gold mineralisation in len-
ticular to pod-like bodies hosted in permeable breccias (stratiform). Minor chalcopyrite, ten-
nantite and stibnite represent late mineralization phases, and gold occurs in association with tel-
lurides.

iii) Low sulphidation pyrite- and base metal-quartz veins which post date the high
sulphidation mineralization.

Emplacement of the hydrothermal breccia and the Imbanguila dacite-diatreme dome complex are
probably related to emplacement of a high level intrusion at depth beneath, or along the margin
of the FSE quartz diorite. Exsolution of reactive magmatic volatiles, and later mineral-izing
fluids from this same high level intrusion at around 6.5-7 m.y., have produced the zoned
advanced argillic alteration and copper-gold-arsenic mineralization in the various settings out-
lined above. It is postulated that there may have been two upflows of high sulphidation fluids,
one associated with emplacement of the hydrothermal breccia and another

92
Fig. 6.25

Fig. 6.26
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

at depth, to anhydrite-alunite (± diaspore, pyrophyllite, zunyite and illite) at shallow levels. This
advanced argillic alteration extends beyond the breccia and forms and extensive cap over the
FSE porphyry. The alteration grades laterally from: quartz-alunite-zunyite in central zones,
through pyrophyllite-diaspore, to margins of illite-chlorite (T. Leach, unpubl. data). The alu-nite
in the hydrothermal breccia has been dated at 6.9-8.5 m.y. (Hedenquist, unpubl. data).

iii) The funnel shaped Imbanguila dacite porphyry has variably been dated at 5.6-7 m.y. (Gar-cia,
1991) and is associated with a diatreme breccia which mushrooms at shallow levels to the SW
(Fig. 6.18). Advanced argillic alteration and associated gold-copper mineralization prefer-entially
occur at the intersection of faults and fractures (of the Lepanto Fault) and the contact between the
diatreme and the underlying volcaniclastics. Central zones of vughy to massive quartz grade
outward through quartz-alunite-kaolinite, and kaolinite zones to peripheral smec-tite-illite and
chlorite alteration. The dacite porphyry is relatively unaltered except at its mar-gins and in fault
controlled quartz-pyrite-anhydrite zones (Easterlies) grading out to quartz-alunite-kaolinite
alteration.

iv) The Bato Dacite forms porphyry domes and associated tuffaceous diatreme breccias, and
yielded a K-Ar age on biotite of 2.9 m.y. (Sillitoe and Angeles, 1985). Garcia (1991) interprets
these dacite units to be post-mineral. To the west of the mine area, quartz-alunite-kaolinite al-
teration occurs as a 6 km long ridge which defines the lower contact of dacitic units. Alunite in
these ridges (Hedenquist, unpubl. data), as well as sericite in the FSE, have been dated at
between at 1.4-3.3 m.y.

Three main types of mineralization are recognised (Garcia, 1990, 1991) at Lepanto -FSE:

i) Porphyry style chalcopyrite-bornite mineralization hosted within and adjacent to the FSE
quartz diorite is related to the chlorite-illite/sericite-clay overprinting alteration event. Gold
occurs as inclusion in, and overgrowing, copper sulphides. This mineralization is interpreted to
have been either deposited during, and/or remobilized by, the post-FSE hydrothermal breccia
event.

ii) High sulphidation enargite-luzonite mineralization is associated with advanced argillic alter-
ation and hosted within central vughy to massive quartz zones: in alunite-anhydrite zones with-in
the hydrothermal breccia, at the diatreme breccia - metavolcanic contact along the strike of the
Lepanto Fault (branch veins or classical ore), in fault controlled alunite-anhydrite zones which
cut the dacite porphyry (Easterlies), and in flat lying bodies which replace calcareous sediments
(stratabound ore). Kaolin clay alteration hosts copper-gold mineralisation in lenticu-lar to pod-
like bodies hosted in permeable breccias (stratiform). Minor chalcopyrite, tennantite and stibnite
represent late mineralization phases, and gold occurs in association with tellurides.

iii) Low sulphidation pyrite- and base metal-quartz veins which post date the high
sulphidation mineralization.

Emplacement of the hydrothermal breccia and the Imbanguila dacite-diatreme dome complex are
probably related to emplacement of a high level intrusion at depth beneath, or along the margin
of the FSE quartz diorite. Exsolution of reactive magmatic volatiles, and later mineral-izing
fluids from this same high level intrusion at around 6.5-7 m.y., have produced the zoned
advanced argillic alteration and copper-gold-arsenic mineralization in the various settings out-
lined above. It is postulated that there may have been two upflows of high sulphidation fluids,
one associated with emplacement of the hydrothermal breccia and another

92
Fig. 6.25

Fig. 6.26
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

along the contact of the dacite-diatreme complex. Fluids have outflowed along the NW-
trending dilatant structures. Some copper mineralization at Lepanto may have been remo-
bilized from an early porphyry copper event at FSE; and some gold may have been depos-
ited at FSE from the high sulphidation fluids.

Garcia (1991) interprets the base metal-gold mineralization to be related to neutralization of

this same hydrothermal system during late stages of activity. The advanced argillic ridges to
the west of the mine are interpreted (Garcia, 1991) to be post mineral and associated with a
more recent hydrothermal event.

iii) MLXasi, Fiji

The Mt. Kasi Prospect, Fiji (Fig. 1.2) represents a structurally controlled style of high sul-
phidation alteration and gold (copper) mineralization, formed at a high crustal level and only
poorly eroded. This discussion is taken from Corbett and Taylor (1994) and Leach (unpubl.
report, 1994). At the time of writing (early 1996) a construction is in progress of a mine which
will extract a resourse comprising: 1,048,000 t of eluvial Au @ 1.9 g/t and 1,240,000 t of hard
rock @ 3 g/t Au (total of 180,000 oz Au). Workings form 1932 to 1948 produced 261,000 t @
7.5 g/t Au (63,000 oz Au). Mt Kasi is a high level gold-rich and copper-poor high sulphidation
system.

Host rocks comprise Late Miocene lavas and pyroclastics which are intruded by dacite domes.
An aeromagnetic high in the vicinity of the Mt. Kasi Prospect may represent a magnetite-
bearing altered intrusive at depth and appears to be offset with a sinistral displacement, by
NNW trending corridor of structures termed the Mt. Kasi Fault System (MKFS). The high
contrast between the resistive silicification and the enclosing conductive clay alteration has
facilitated the subsurface mapping of the Mt. Kasi alteration system by CSAMT (controlled
source audiomagnetotelluric; Smith and Irvine, 1990) geophysics (Fig. 6.27; Corbett and Tay-
lor, 1994).

Zoned alteration extends from locally steeply-dipping silicification as upflow zones, laterally
into outflow features, which were found to be rootless in early drilling. Upflows tend to be
localised by intersections of cross structures with the MKFS and palaeo fluid flow directions
are apparent from the shapes of upflow-outflow relationships (Figs. 6.27, 6.28). Limited data
suggest that outflows might radiate from a central upflow in the vicinity of a dilational jog in
the MKFS (Fig. 6.27). The cross faults may provide late- to post-mineral offsets of the alter-
ation in a configuration similar to domino structures (Section 3.vi). The 1100 workings are
therefore inferred to represent an upflow to an outflow in the vicinity of the open pit. Other
upflow-outflow relationships are apparent at Done Creek and Kasi South (Fig. 6.27).

The overprinting of alteration by mineralization is apparent at Mt. Kasi. The individual fluid
upflow-outflow centres derived from the initial vapour-dominated fluid display an alteration
zonation grading outwards as assemblages dominated by: a core of residual (vughy) silica, sil-
ica alunite, pyrophyllite and peripheral kaolin. Gold and copper mineralization associated with
the later liquid-dominated fluid exploited the same plumbing system during continued
deformation on the MKFS. Ore forms the matrix to breccias within the competent residual
silica and gold (copper) grades are proportional to the matrix content of the breccias. Matrix
supported rotational breccias proximal to the fluid upflow zones contain higher gold grades
than the peripheral fluidised and crackle breccias in the outflow zones (Corbett and Taylor,
1994; Fig. 6.28). NW trending fractures, slickensided faults and sigmoidal-shaped fluidised
breccia zones represent dilational ore-hosting features, and are indicative of a continued sin-
istral rotation along the MKFS (Fig. 6.28; Corbett and Taylor, 1994).

93
Fig. 6.27

Fig. 6.28
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Alteration at Done Creek defines concentric zones grading outwards from: a core of vughy s il-
ica in which cavities are filled by kaolinite ± dickite, to assemblages of quartz-kaolinite-
interlayered, illite-smectite and peripheral sub-propylitic chlorite-carbonate. Local alunite in-fills
leached vughs and has been replaced by later kaolinite.

Breccia-hosted gold mineralization consists of an early phase dominated by pyrite-quartz and a

later phase of gold-copper mineralization. Ore phases have been deposited in open fractures and
leached vughs adjacent to fractures and overgrow earlier quartz and barite. Vertical gradations in
mineralogy include: luzonite-tennantite-chalcopyrite at depth to goldfieldite-tennantite at shallow
levels (including float boulders). Lateral gradations to chalcopyrite-tennantite-galena-sphalerite in
peripheral argillic zones also occur elsewhere at Mt Kasi (Turner, 1986). Bonanza mineralization
(locally > 1 percent Au) in float boulders at Done Creek occurs as high fineness (>900) native
gold, deposited as inclusions in, and overgrowing, tennantite and goldfieldite, overgrowing pyrite,
and infilling vughs in earlier quartz-pyrite veins. Trace gold-tellurides (mainly calaverite) occur
as minute inclusions in goldfieldite and tennantite. Inclusions of copper-tin sulphide phases
(colusite and hemusite), which contain appreciable vanadium and molybdenum contents
respectively, occur in some high grade silicified float.

Mt. Kasi is a high sulphidation system which is exposed at very shallow, epithermal levels based
on the dominance of quartz-kaolinite-dickite as the main alteration phases, luzonite-tennantite-
goldfieldite as the copper ore phases, and low homogenisation temperatures in bar-ite within
mineralized zones (averages of 165-220°; Turner, 1986). The association of bonanza grade gold
mineralization with tellurium (± vanadium) in this epithermal high sulphidation sys-tem is
comparable to the bonanza grade deposits in low sulphidation, intrusive-related, epi-thermal
systems (e.g., Zone VII at Porgera).

Fluid flow models are apparent on outcrop and prospect -scale. Individual fluid upflow-outflow
centres vector away from the central portion of the hydrothermal system where a fault jog is
inferred from the sinistral rotation on the MKFS (Fig. 6.26). Outcrops of dacite here could be
indicative of a magmatic source.

v) Composite Structurally and Lithologically Controlled Gold-Copper High Sulphidation

Systems

a) Characteristics

Most high sulphidation gold-copper systems display aspects of both lithological and structural
control and those categorised above as lithologically or structurally controlled are in essence end
members of a continuum. Composite controls are evident within different portions of the same
system or as changes with time. A diatreme margin could be classed as a permeable lithological
contact by some workers or structural contact by others. Dilatant structures which tap the
magmatic source, typically control the fluid flow at depth. Upon contact with permea-ble host
rocks a lithological control may be evident, particularly in the upper portions of many systems.
One high sulphidation system may demonstrate structural control in some portions and
lithological control in others.

Systems which display approximately equal structural and lithological control are Maragorik,
East New Britain, PNG; (Corbett et al., 1991; Corbett and Hayward, 1994); Peak Hill, eastern
Australia; (Degeling et al., 1995); Bawone-Binebase, Sangihe Is, Indonesia (Corbett unpubl.

94
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

report, 1993); Temora (Thompson et al., 1986) and Dobroyde (Leach, unpubl. data) in eastern
Australia.

b) Examples

i) Peak Hill, Eastern Australia

Although occurring within an Ordovician magmatic arc of the Lachlan Fold Belt, eastern Aus-
tralia (Walshe et al., 1995), Peak Hill displays features typical of younger high sulphidation
gold-copper systems as summarised here from Degeling et al., (1995). The inferred magmatic
source for the high sulphidation alteration and mineralization may have been localised by the
intersection of NW trending transfer structures which offset the magmatic arc, and the arc -
normal Parkes Thrust. Host rocks comprise andesitic volcanics and epiclastic rocks.

An initial lithological control to the high sulphidation alteration is evidenced by the localisa-tion
of silicification at the intersections of NW trending structures and permeable host rocks (e.g.,
Bobby Burns workings, Fig. 6.29). These structures also provide post -mineral offsets to the
alteration (e.g., Crown workings, Fig. 6.29), and host possible earlier low sulphidation quartz
veins. In addition, NW structures localise fracture controlled mineralization which is best
developed in portions of the NW structures which deviate to WNW trends (e.g., Proprietary open
pit, No. 2 and Mingelo stopes, Fig. 6.29). The model presented by Degeling et al. (1995)
suggests that regional sinistral rotation, possibly on arc-parallel structures such as the Gilmore
Suture (Stuart-Smith, 1991), has facilitated the formation of local mineralized WNW-trending
jogs where the NW fractures transgress the competent silicification.

Four distinct stages of hydrothermal activity have been recognised at Peak Hill (Fig. 6.30):

Stage I: Quartz veins host gold mineralization at Myall United or McPhails workings north of
Peak Hill, and at the Crown workings at Peak Hill, and predate the high sulphidation minerali-
zation. These veins strike NW and exhibit locally higher grades in WNW-trending jogs.

Stage II: This is the main high sulphidation system which developed progressively as follows:

i) An early phase of alteration exhibits a pronounced lithological control in the exploitation of

favourable permeable units in the epiclastic/volcanic sequence, over an area of 500 x 1000 m. At
Proprietary (Fig. 6.31) the alteration is zoned outwards from a central core of residual vughy to
massive quartz which is hosted in steeply dipping fine grained pyroclastic and rimmed by silica-
alunite. The silicified zones grade to silica-micaceous clay alteration which is broad to the east
and narrow to the west. The micaceous clays are interpreted to have been re-crystallised during a
post-high sulphidation deformation event, and grade from sericite at depth and in the south, to
pyrophyllite at shallow levels and to the north. Trace andalusite co-exists with pyrophyllite at
Great Eastern. Silica-paragonite, paragonite-chlorite, chlorite-albite and epidote-albite-chlorite
alteration form as progressive zones westward of the silica-micaceous clay alteration, and are
hosted in less permeable andesite volcanics. The zoned al-teration is interpreted to reflect
progressive neutralization and cooling of a hot acid fluid as it migrates away from permeable
lithologies. The extent of silicification is greatest closer to the inferred NW feeder structures and
dies out moving along the strike of the permeable units, (e.g., at Parkers; Fig. 6.29).

ii) The zoned alteration, and especially the more brittle quartz and quartz-alunite zones, have

95
Fig. 6.29

Fig. 6.30
Fig. 6.31

Fig. 6.32
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edti.

undergone fracturing and local brecciation, accompanied by deposition of quartz-barite ± alu-

nite. In drill hole OPH2, south of Peak Hill, silicified vughy volcanics have undergone intense
fracturing and brecciation, and are sealed in a vein breccia of coarse tabular alunite. This style of
alunite vein/breccia is common in high sulphidation systems proximal to source intrusions (see
section 4.ii.b).

iii) Further fracturing and brecciation was accompanied by deposition of sulphide phases which
comprised of earlier massive pyrite, followed by later deposition of copper -gold ore phases.
Copper-gold mineralization at Proprietary is localised at the intersection of the central residual
silica zones and the NW trending feeder structures. Sub-economic copper mineralization at
Proprietary, and to a lesser degree at Parkers, is restricted to a quartz-pyrite-barite zone, and is
dominated by tennantite and minor luzonite. Tennantite is locally enriched in tellurium, and trace
minute Au-tellurides (calaverite) have been reported as inclusions in some pyrite (Alli-bone,
1993). High fineness (943-968) native gold occurs with tennantite infilling fractures cut-ting
pyrite. The occurrence of Te-rich mineralogy, tennantite-luzonite copper mineralization, and free
gold are indicative of shallow epithermal levels in a high sulphidation system. Chalco-pyrite-
enargite ± bornite mineralization in the southern region of Bobby Burns implies higher
temperature mineralization there, than in prospects to the north.

Stage III: This is the main phase of post-alteration-mineralization deformation and shearing.
The zonation in micaceous clay outlined above is inferred to indicate lower pH conditions in
the north during deformation.

Stage IV: Late stage deposition of kaolinite and gypsum, and at depth fine grained pseudo -
cubic alunite in open cavities and breccia zones, implies a collapse of cool, acidic fluids onto
earlier alteration assemblages. In places the pseudo -cubic alunite is slightly deformed, howev-er
in most cases undisturbed, indicating that most of the Stage IV retrograde activity was post
deformation/shearing.

Information from structure, alteration and mineralization suggest that hot acidic magmatic flu-
ids have been derived from an intrusive source in the vicinity of a magnetic high about 1.5 km
to the southeast of Peak Hill. It is interpreted that volatile-rich magmatic fluids migrated along
the arc normal NW structures, and caused zoned alteration centred in permeable pyroclastic
units. Later mineralized fluids have moved north and east along the same regional structures,
and deposited gold-copper mineralization along WNW-EW trending fractures hosted in brittle
silicified zones.

ii) Maragorik, East New Britain, Papua New Guinea

The Maragorik Prospect, East New Britain, Papua New Guinea (Fig. 6.32), is a high sulphida-
tion gold-copper system which have undergone only minor erosion (Corbett et al., 1991; Cor-bett
and Hayward, 1994). As extensive ash deposits blanket the region, CSAMT geophysics in
conjunction with bulldozer trenching, were utilised to delineate the subsurface geology. At
deeper levels fluid upflow occurred along EW structures dilated by the rotation on the bound-ing
major NW structures (Fig. 6.33). At higher levels, the rising hydrothermal fluids have ex-ploited
permeable horizons which intersect the upflow structures and demonstrate a lithological control
to form ledges of silicification and peripheral clay alteration (Fig. 6.34). Thus, zones of
silicification occur as steep and flatly dipping ledges. As is typical of high sulphidation sys-tems,
an initial inferred vapour-dominated phase is followed by a liquid-dominated phase. Much of the
zoned silica to clay alteration is developed during the early phase of activity. Mineral deposition
occurs as a result of brecciation of the competent

96
Fig. 6.33

Fig. 6.34
silicification by later phase fluids and is restricted to the ledges proximal to the feeder
structures.

Styles of alteration and mineralization are indicative of a very low temperature and hence high
level system, characterised by opaline silica, smectite dominated clays and luzonite as the low
temperature polymorph of enargite. Although high sulphidation systems are inferred to develop
from porphyry-related magmatic fluids, such a source at Maragorik is interpreted to be very
deeply buried.

iii) Bawone-Binebase, Sangihe Island, Indonesia

At Bawone-Binebase on Sangihe Island, Indonesia both structurally and lithologically con-

trolled high sulphidation gold-copper mineralization are interpreted to have been derived from
the one fluid source and occur within different parts of the same hydrothermal system (Fig.
6.35; Corbett unpubl. report, 1993). Low grade porphyry alteration and mineralization occur at
Binebase and elsewhere on Sangihe Island. Low sulphidation mesothermal quartz-sulphide
veins are inferred to represent the hypogene source for supergene gold recovered by illegal
miners at Taware Ridge. The inferred magmatic source for the high sulphidation system is lo-
calised on the margin of a NNW graben by the intersection of thoroughgoing NNE structures,
and dilation of ESE structures by sinistral rotation on NNW structures (Fig. 6.35). The inter-
secting structures have tapped the magmatic source forming a fluid upflow feature.

At Bawone, a fluid flow model is apparent from zoned alteration and gold-copper distribution in
several cross sections (Fig. 6.35). Hot magmatic fluids are inferred to have been derived from the
vicinity of overprinting diatreme breccias and flowed laterally along the dilatant struc-tures
towards the SE. The size of the alteration zones, temperature of formation and metal grades all
decline moving from the upflow to outflow settings. Zonations and paragenetic se-quences of
overprinting alteration and mineralization are typical of high sulphidation systems. The local
sharp contacts between: residual silica, silica-alunite and peripheral clay alteration, are indicative
of a high level setting or distal relationship to the inferred magmatic source, and typical of an
outflow portion of the hydrothermal system. Mineralization occurs as sulphide-rich matrix to
fluidised breccias and sulphide infill of vughs in the competent altered residual silica and silica-
alunite.

While the bulk of the hydrothermal fluids have flowed to the SE along the dilatant structures,
relatively small structurally controlled high sulphidation mineralization occurs to the SW at
Brown Sugar and Bonzo's Salvation. Rapid changes in alteration zonation are consistent with
fluid quenching and low temperature clays are also indicative of the dilational setting.

The Binebase alteration resulted from the northward migration of hydrothermal fluids along a
corridor provided by the intersection of a permeable lapilli tuff unit and thoroughgoing NNE
structures. Low temperature alteration assemblages are consistent with the distal relationship to
the inferred fluid source at Bawone. Chalcedony becomes increasingly vughy down dip and to
the south towards the inferred upflow. As seen in some other lithological controlled systems,
there is little distinction between alteration and mineralization resulting from the vapour-
dominated phase I fluid, and the later liquid-dominated stage II mineralized fluid. The abun-
dant gypsum and barite also suggest that incursion of seawater could have occurred, possibly
from the NW.

97
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

vi) Hybrid High-Low Sulphidation Systems

a) Characteristics

Giggenbach (pers. commun., in Hedenquist, 1987) states that "ascent of volcanic (magmatic)
gases and their transition from an oxidised (sulphur as SO2 - high sulphidation) to reduced
(sulphur as H2S - low sulphidation) state is 'a battle of the buffers', in which each achieves a
partial victory". Hedenquist (1987) postulated that there is a continuum from high to low sul-
phidation systems, which is dependent on the degree of access of these upwelling fluids to neu-
tralization (and cooling) through reaction with the wall rock and/or circulating surficial waters.

All high sulphidation systems exhibit zoned alteration, which is indicative of this process of
cooling and neutralization within subsidiary structures or permeable lithologies. In this envi-
ronment the magmatic-derived fluids are able to be modified away from the major feeder
structures. However, in certain cases the upwelling hot acidic, magmatic-derived high sul-
phidation fluid becomes cooled and neutralized within the major regional structures themselves.
This results in a transition from high to low sulphidation type fluid and the formation of a hy-brid
deposit type (e.g., Wild Dog, PNG). Elsewhere, the initial hydrothermal fluid may be dom-inantly
high sulphidation, but a later fluid may be low sulphidation in nature. This reflects changes in the
chemistry of the fluids which exsolve from the magmatic source during late stage of melt
crystallisation, or the mineralized fluid has been modified during its ascent. A superimposed high
and low sulphidation system might be the base metal gold veins which cut the high sulphidation
system at Lepanto, Philippines (section 6.iv.b), or the banded epithermal quartz veins which cut
advanced argillic alteration at Masupa Ria, Indonesia (Thompson et al., 1994).

b) Examples

The enigmatic Wild Dog Prospect, Papua New Guinea (Lindley, 1987, 1988, 1990) displays
characteristics of both high and low sulphidation gold systems, and Arribas (1995) notes that
Masupa Ria, Indonesia (Thompson et al., 1994) and the Kelly mine, Philippines (Comosti et al.,
1990) are examples of possible transitions from high to low sulphidation systems.

i) Wild DogT East New Britain, Papua New Guinea

The Wild Dog Prospect was identified in 1983 during the follow up of anomalies including al-
tered float and pannable gold identified during a regional stream sediment exploration pro-
gramme (Lindley, 1987). Evaluation of the project by Esso (PNG), City Resources and High-
lands Gold Limited continued until the early 1990's. Host rocks comprise andesitic to dacitic
lavas and tuffs to which Lindley (1987, 1988) attributes a probable Mio-Pliocene age. Recent
ash partly blankets the area.

Wild Dog is one of several alteration zones hosted within the Warangoi Structural Corridor,
which transects an inferred Nengmutka caldera (Lindley, 1987, 1990). The caldera is localised
within the Baining Mountain Graben structures, which data showing the depth to the mantle
(Wiebenga, 1973), may represent the margin of a deep rift (Fig. 6.36, Corbett, unpubl. report,
1990). At prospect scale three NNE trending and west dipping silicified zones occur within the
Warangoi Structural corridor as a prominent ridge (Lindley, 1990). NW trending cross struc-
tures exploited by the drainage pattern and locally offset the silicified zones as slickensided
faults (Corbett unpubl. report, 1990, Fig. 6,37).

98
Fig. 6.36

Fig. 6.37
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Two main hydrothermal events are recognised in the prospect area (Fig. 6.38):

i) Replacement silicification of regionally propylitic altered (epidote-pyrite-chlorite) volcanics,

produced a dense, grey fine grained chert-like alteration (Lindley, 1990). These steeply dip-ping
silicified zones are unmineralized, pinch and swell up to true widths of 50-70 m, and are aligned
NNE parallel to the Warangoi Structure. Alteration mineralogy in the silicified zones and the
immediate wall rock is vertically zoned at Wild Dog from: sericite ± pyrophyllite at depth,
through sericite, to local sericite ± chlorite at shallow levels. Trace molybdenite miner-alization is
associated with the silicification event.

Similar structurally controlled silicification is locally encountered along the Warangoi Struc-
ture at Keamgi Hill, 2 km SSW of Wild Dog, and Kasie Ridge, 4 km to the NNE (Fig. 6.36).
At Kasie Ridge, 300 m lower elevation than Wild Dog, subparallel NNE trending silicified
ridges are zoned from: central zones of quartz-alunite ± zunyite ± pyrophyllite + diaspore,
through pyrophyllite-sericite ± kaolinite/dickite and sericite ± illitic/kaolin clay, to peripheral
chlorite—illitic clay, which grades outwards to regional propylitic alteration. This zonation is
comparable to high sulphidation systems encountered elsewhere in the southwest Pacific.

ii) Polyphasal quartz tension gash veins transect the silicified zones, commonly as hanging wall
splits, best developed near the cross structures (Lindley, 1990; Corbett unpubl. report, 1990).
Later mineralization infills open fractures and cavities in the quartz veins, and forms dark bands
containing copper mineral phases (chalcopyrite and minor bornite, chalcocite and tennantite),
and local Cu-Bi-Pb-Ag sulphides, tellurides and sellenides. Gold is generally re-stricted to Au-
Ag telluride phases (Lindley, 1990), and native 'mustard' gold occurs as an al-teration
(weathering) product of these tellurides.

Zonations in illitic and smectitic clays and fluid inclusion studies in the late quartz veins sug-gest
that copper-gold mineralization took place in response to the mixing of cool (<200°C) and dilute
(<2.0 wt percent NaCl) meteoric waters, with upwelling hot (>280°C) and saline (>15 wt percent
NaCl) fluids.

It is interpreted that the prospects in the Wild Dog region are composite high- and low-
sulphidation systems. Initial silicification was derived from hot acidic fluids which exsolved
from a crystallising high level melt into the Warangoi Structure (Fig. 6.36). These acidic flu-ids
progressively became neutralized at shallower levels as indicated by the zonation from alu-nite-
zunyite-pyrophyllite at Kasie Ridge, through pyrophyllite and sericite, to near surface se-ricite-
chlorite at Wild Dog and Keamgi Hill. This is comparable to the initial vapour-rich leaching
event in high sulphidation systems.

Late stage mineralization is hosted in fractured silicified zones and inferred to have been de-
rived from a contemporaneous release of magmatic mineralized fluids from depth (e.g., the
parent melt). These fluids mixed with cool dilute meteoric waters within local tension gash
structures resulting in the Cu-Bi-Pb-Te-Au mineral deposition which is typical of low sul-
phidation quartz-sulphide lodes. However, the common occurrence of tetrahedrite and chalco-
cite is indicative of fluid conditions which are transitional to a high sulphidation type.

High and low sulphidation systems are differentiated on the basis of fluid chemistry (Section
l.iii), i.e., whether sulphur SO2 (high sulphidation) or H2S (low sulphidation) is predominant as
the main dissolved sulphur gas phase. In high sulphidation systems the upwelling hot acidic
fluids are confined within major regional structures. However, these fluids are progressively
neutralized and cooled within subsidiary structures or permeable lithologies, where they have

99
Fig. 6.38

Fig. 6.39
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

the opportunity to react with the wall rock and/or mix with neutral circulating surficial waters,
and form zoned advanced argillic —> argillic —> propylitic alteration assemblages. Later
mineralizing fluids are also typically acidic and deposit metals in the fractured and brecciated
alteration zones. Low sulphidation fluids on the other hand are interpreted to be formed either
through:
* the neutralization and cooling of low pH magmatic fluids at the base of permeable
circulating meteoric systems, or
* as magmatic fluids which are exsolved from the crystallising magma but are low in
dissolved reactive gases.

ii) Masupa Ria, Central Kalimantan, Indonesia

Overprinting low and high sulphidation systems at Masupa Ria have been described by
Thompson et al., (1994) and Leach (unpubl. data). Flat lying ridges of zoned silica and
advanced argillic alteration at Masupa Ria extend for up to 7 km within northeast and
northwest regional structures (Fig. 6-39). These ridges consist of massive to vughy silica
which grade with increasing depth through pyrophyllite-kaolinite-dickite and intense
quartz-sericite alteration to regional epidote-chlorite-calcite propylitic alteration. This
alteration is hosted in flat lying pyroclastic units which are interpreted to have acted as
permeable host rocks for outflowing high sulphidation-style acidic fluids.

Although barren of mineralization, the silica-alunite ridges have locally acted as brittle host
rocks and fractured to host later low sulphidation style vein-mineralization. The Ongkang vein
system trends parallel to northwest regional structures, and swells at the intersection with Ma-
supa Ria silica ridge. Veins consist of colloform banded quartz (locally after bladed car-bonate),
typical of intrusive-related low sulphidation gold-silver quartz vein systems formed at
epithermal levels (see section 7.iv). Fluid inclusion analyses indicate that coarse quartz was
deposited at 250-300°C and under dilute (<3 wt percent NaCl) conditions. Mineralization is
restricted to thin sulphide bands composed of fine quartz, low temperature illitic and chlorite
clays, rare base metal sulphides and trace silver sulphosalts and sulphides. Gold occurs as mi-
nute free grains intergrown in the sulphide bands and in fractures cutting the banded quartz, and
exhibits an average fineness of around 820 (see section 7.i). Fractures and cavities are in-filled
by barite, gypsum, kaolinite and smectitic clays. Gold mineralization is interpreted to have
occurred in response to the mixing of hot mineralized fluids with cool meteoric waters.

It is not clear whether the silica ridges formed by high sulphidation fluids, and the auriferous
quartz veining deposited by low sulphidation fluids, are different phases of the same magmat-
ic-related hydrothermal system, or completely separate overprinting systems.

100
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

7 PORPHYRY-RELATED LOW SULPHIDATION GOLD SYSTEMS

i) Classification

a) Introduction

The emplacement of intrusions in magmatic arcs provides heat sources to drive deep circulating
hydrothermal systems which are composed of varying proportions of magmatic and meteoric
fluids. Magmatic volatiles which evolve from the cooling magma, become entrained at the base
of the circulating fluids. Rock reaction and fluid mixing reduce acidic gases such as SO2 to
H2S, and Cl" to dissolved salts, mainly NaCl. As outlined in section l.iii, in these systems sul-
phur is present at an oxidation state of -2, and so this style of hydrothermal system is classi-
fied as low sulphidation (Hedenquist, 1987).

In crustal settings where the intrusions are emplaced into thick permeable volcanic piles, the
magmatic fluids become diffused within large circulating hydrothermal systems. This diffusion
results in the formation of broad alteration zones, characteristic of many active porphyry-related
geothermal systems (Section 2.iii), and commonly encountered in porphyry-copper sys-tems
(Section 5). Regional structures provide enhanced permeability which channel outflow
considerable distances from the intrusion heat source (e.g., 15-20 km at Bacon-Manito, Phil-
ippines).

However, in magmatic arcs at continental margins, high level porphyry intrusives are em-
placed into impermeable host rocks such as older plutons, sediments and metamorphic base-
ment rocks (see Section 2.iii.c). In these environments circulating hydrothermal fluids migrate
along zones of permeability in competent host rocks provided by structures (e.g., dilational jogs
or splays in major structures, sheeted fractures, and fracture permeability), breccias (e.g.,
diatreme margins, intrusive fluidised breccias), or geological contacts (e.g., fractured dome or
dyke margins). This focusing of hydrothermal fluids provides an ideal geological and hydro-
logical environment for the formation of porphyry-related gold systems.

b) Sequence of Events

Low sulphidation systems in the southwest Pacific region exhibit similar paragenetic sequenc-es
of events at all crustal levels, and this is interpreted to reflect the interaction between melts
emplaced at high crustal levels and circulating meteoric waters (Fig. 7.1). The transfer of heat
from a melt emplacement at high crustal levels, causes heating of waters residing in structures
and permeable units and thereby produces vertically zoned fluidised breccias, diatreme-maar
complexes, and pebble dykes, which utilise pre-existing structures (see section 3). The em-
placement of melts also provides the heat source for the development of a convective hydro-
thermal system which is dominated by cool dilute meteoric waters. The hydrothermal system
circulates to depths along regional and subsidiary structures, intrusive contacts and/or permea-
ble lithologies. These circulating meteoric-dominated waters deposit vein systems dominated by
quartz and secondary K-feldspar, commonly adularia (Henley and Ellis, 1983). Magmatic
volatiles, which progressively evolve from the melt as it crystallizes, are entrained within the
meteoric waters at the base of the convecting hydrothermal system and become neutralized and
reduced as outlined above. CO2 is commonly the dominant magmatic volatile phase (with the
obvious exception of water vapour) and is the only major magmatic component which is little
effected by secondary processes, although some is converted to methane at low temperatures
(Giggenbach, 1987). The increase in dissolved gases lowers the pH of the circulating waters and
favours the deposition of sericite/illite over

101
Temporal and Spatial Zonations in Low Sulphidation Gold Systems
Fig. 7.1

Fig. 7.2
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

K-feldspar (Browne, 1991) in progressively later quartz veins.

As the intrusion continues to cool, pressure drops result in the progressive draw -down of near
surface bicarbonate condensate and acid sulphate waters to depths of up to 1.5 -2 km into the
active hydrothermal system (Reyes, 1990a; Mitchell and Leach, 1991). At cool epithermal lev-
els the downflowing acid sulphate, bicarbonate and dilute groundwaters are oxygenated relative
to the deeper circulating hydrothermal fluids (Robinson et al., 1987).

Crystallization of the intrusion and exsolution most of the volatile phase precedes the late stage
segregation of metals from the melt (Cline and Bodnar, 1991). Metal-bearing magmatic fluids
are released into structures (and other permeable channelways) which are interpreted to
contain:
* circulating cool dilute meteoric waters at all levels in the system,
* dilute low pH gas condensate ± acid sulphate waters up to 1.5-2.0 km depth (but
predominantly at >1 km depth),
* relatively oxygenated waters at shallow epithermal levels.
Mineralization at all crustal levels characteristically postdates the quartz-K-feldspar-
sericite/illite veining (see following examples). Iron sulphides are overprinted by base and
precious metal phases. This is also a common feature of high sulphidation systems (section
6), and is interpreted to possibly reflect an earlier partitioning from the crystallising melt of
iron, relative to base and precious metals.

As the hydrothermal system continues to wane, surficial fluids descend to deeper levels in the
hydrothermal system, and result in low temperature clay alteration and carbonate-sulphate
deposition, which overprint earlier assemblages.

c) Types of porphyry-related low sulphidation gold systems

Much of the early geological literature arising from studies in porphyry copper terrains has
applied the term "epithermal" to describe porphyry-related gold deposits formed outside the
porphyry environment. During the upsurge of gold exploration in the 1980s, it became diffi-
cult to place many southwest Pacific gold deposit types in the existing classification. The in-
creased data facilitated comparisons of different deposits, and so the formerly unique Porgera
gold deposit (Sillitoe, 1989), has more recently been grouped as part of the carbonate-base
metal gold classification (Leach and Corbett, 1994, 1995; Corbett et al., 1995). The group of
gold deposits formerly described as epithermal are subdivided as porphyry-related low sul-
phidation gold deposits according to the crustal level of formation and relationship to porphyry
source (Leach and Corbett, 1995). These vary from: deepest porphyry levels, to in-termediate
or mesothermal depths (quartz-sulphide gold ± copper and carbonate-base metal gold), and
shallow epithermal levels (Figs. 7.1, 7.2). Although telescoping is common, and overprinting
of alteration zonations may locally obscure the boundaries, deposit types (Fig.
7.3) are distinguished as:

Porphyry copper-gold systems form at deepest levels where magmatic-derived mineralized flu-
ids evolve from a cooling magma and deposit metals within the fractured carapace of porphyry
stocks and adjacent competent country rocks (see Section 5).

Quartz sulphide gold ± copper vein systems form at shallower levels peripheral to porphyry
copper-gold intrusions. The magmatic-dominated fluids are channelled into permeable fea-
tures (e.g., sheeted fractures, structures, or magmatic hydrothermal breccias) where they peri-
odically mix with deep circulating meteoric waters (Fig 7.2). The cooling and dilution

102
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

during mixing results in metal deposition within quartz-sulphide vein/fracture systems. Gold
mineralization is commonly hosted in massive pyrite/arsenopyrite-chalcopyrite ± magnet-
ite/hematite veins, which postdate quartz veining. The early stage quartz veins are deposited from
either dilute circulating meteoric- or saline magmatic-dominated fluids. Lead-zinc phases are
subordinate to copper phases.

Carbonate-base metal gold systems form in response to the mixing of upwelling fluids, contain-
ing a significant magmatic component, with descending bicarbonate-sulphate fluids derived from
surficial gas condensate zones (Leach and Corbett, 1994, 1995). The carbonate-base metal gold
systems are encountered at shallower levels, and/or in later vein sequences, than the quartz-
sulphide gold ± copper mineralization. Gold mineralization is associated with pyrite-sphalerite-
galena and carbonate veins, or breccia infill, with only minimal copper mineralization.

Epithermal quartz gold-silver systems contain precious metals which are inferred to have been
deposited by the mixing of upwelling mineralized fluids containing (diluted) magmatic signa-
tures, with descending oxygenated groundwaters. On the other hand, much of the gangue miner-
alogy comprising quartz, adularia, and quartz pseudomorphing platy carbonate forms in re-sponse
to the boiling of dominantly meteoric fluids upon periodic, structurally-controlled pres-sure
release, and so develop as characteristically banded fissure vein systems. Epithermal quartz gold-
silver systems are encountered at shallow or epithermal levels, and are locally contempo-raneous
with and/or post date carbonate-base metal gold systems. Many of these systems con-form to the
adularia-sericite epithermal gold-silver deposits described in the geological litera-ture (e.g.,
Henley and Ellis, 1983) for which the term adularia-sericite epithermal gold-silver is retained
herein. We have applied the term epithermal quartz gold-silver to those epithermal deposits which
more clearly demonstrate a magmatic association than the adularia-sericite epi-thermal gold-silver
systems. It is anticipated that as our understanding increases with time, these two groups will
coalesce.

ii) Quartz-Sulphide Gold ± Copper Systems a)

Introduction

Quartz-sulphide vein systems are encountered peripheral to porphyry copper-gold intrusions

(Lowell and Guilbert, 1970; Sillitoe and Gappe, 1984) and are locally mined as small scale
deposits e.g., veins peripheral to Bingham Canyon (Babcock et al., 1995). It is speculated that
may also be a genetic link with unseen porphyries (in distal settings) in the formation of some
structurally controlled gold deposits in older terrains such as the Motherlode deposits (Weir and
Kerrick, 1987), some ironstone-hosted gold (e.g., Tennant Creek, Australia, Huston et al., 1993),
quartz-sulphide reefs at Telfer, Australia (Goellnicht et al., 1989; Dimo, 1990). Alt-hough Slate
Belt gold deposits are generally considered to have been derived from a metamor-phic fluid
(e.g., in eastern Australia Hill End, Seccombe et al., 1993; Victoria, Phillips and Hughes, 1995),
evidence is emerging of stronger magmatic associations at Sofala -Hill End, eastern Australia)
and elsewhere.

Some quartz-sulphide vein systems extend for up to 5 km from the source porphyry such as at
Bingham Canyon, Utah, where mining initially focused upon base metal veins and more re-
cently sediment-hosted gold ores (Peters et al., 1966; Sillitoe 1991b; Babcock et al., 1995).
Bingham Canyon displays metal zonations of typical porphyry copper-molybdenum, to cop-per-
gold skarns, Lead-zinc-sliver skarns and lead-zinc-silver lodes which overlap the

103
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

outer skarns and peripheral sediment hosted gold mineralization (Babcock et al., 1995). Pe-
ripheral veins to granitic intrusions in settings such as Cormwall, England were important his-
torical sources of base metals (Edmonds et al., 1975).

In tropical southwest Pacific rim settings quartz-sulphide veins are worked for gold as small
scale (commonly illegal) local miners such as the Philippines (Mitchell and Leach, 1991), In-
donesia (Tawere Ridge, Sangihe Island, Corbett, unpubl. report) and Papua New Guinea (Ara-
kompa, Corbett et al., 1994b), where gold contents tend to be higher than those in the western
USA and Chile (Sillitoe, 1991b), and commonly display substantial supergene enrichment. Sil-
litoe (1991b) stresses that many porphyry-related quartz-sulphide vein deposits are not of eco-
nomic importance, in terms of the overall gold content. In many Pacific rim tropical settings,
drilling of primary ores often fails to substantiate the apparent rich gold contents won by small-
scale (commonly illegal) miners who gouge narrow structures in the supergene weath-ered
environment.

While quartz-sulphide veins are disappointing exploration targets, in favourable conditions

this style of mineralization may from large bulk low grade deposits for instance: Lihir, PNG
(42 M oz contained Au); in eastern Australia, Kidston (4 M oz Au), Mt Leyshon (>1 M oz
Au), and Cadia 10 M oz Au), Ravenswood-Charters Towers (7 M oz Au) and Lake Cowal (2.4
M oz Au). Others (Sillitoe 1991b) include Zhao-Ye, China (approx. 16 M oz Au) and Kori
Kollo, Bolivia (5 M oz, Au).

Of interest is that analysis of mineral zonations and fluid plumbing systems may point towards
the porphyry source rocks for these systems. Follow up of gold anomalies shedding from pe-
ripheral veins led to the identification of the Batu Hijau porphyry copper-gold deposit, Indo-
nesia (R. Burke, pers. commun.; Meldrum et al., 1994) and porphyry targets are apparent from
the distribution of peripheral quartz-sulphide vein systems at Bilimoia-Arakompa, PNG (be-
low; Corbett et al., 1994b).

b) Structural setting

Mesothermal veins exploit pre-existing structures which tap fluids venting from porphyry
source rocks. These are best developed within dilational structural environments which may
contribute towards the formation of ore-shoots within vein systems. Typical structural settings
which host mesothermal veins peripheral to island arc porphyries are:
* Sheeted vein systems may become prospective by an increase in the density of veins
(Fig. 3.11), (e.g., Kidston, Fig. 7.7) or more importantly, in settings where deformation
enhances the formation of dilational structural environments within the sheeted vein
systems (Fig. 3.12).
* Fluidised and crackle breccias (Fig. 3.16), (e.g., Lihir Island gold deposit, PNG).
* Tension structures parallel to the direction of compression (Fig. 3.9), (e.g.,
Arakompa, PNG, Fig. 7.10).
* Conjugate fractures formed peripheral to porphyry intrusions in orthogonal
compression or extension (Fig. 3.9), (e.g., Batu Hijau, Indonesia, Meldrum et al.,
1994).
* Arc normal structures (e.g., exhumed deep structures host the vein system at
Bilimoia, PNG, Fig. 7.10).
* Dilational tension vein systems (Lake Cowal, eastern Australia),
* Rotation on conjugate structures may form higher grade sigmoidal portions of
conjugate structures (Fig. 3.9), (e,g., Batu Hijau, Indonesia, Meldrum et al., 1994).

104
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

c) Alteration and mineralization

As outlined above the authors interpret that mineral deposition in quartz-sulphide gold ± cop-per
systems results from the mixing of hot mineralized magmatic-dominated fluids, evolved from
cooling intrusives, with meteoric waters, which circulate to considerable depths along ma-jor
regional structures (Fig. 7.2). The quartz-sulphide vein systems exhibit a common parage-netic
sequence of deposition which may be summarised as:

1. Breccias which range from major magmatic hydrothermal eruptions (e.g., Kidston, Australia,
below), through fault-controlled pebble dykes (e.g,, Arakompa, PNG, below), to fluidised
breccias (e.g., Lihir, PNG, below), commonly form as precursors to the hydrothermal system, in
response to the heating of groundwaters by the initial intrusive emplacement.

2. Quartz veins form commonly medium to coarse grained granular to coxcomb textures
associated with pyrite and early K-feldspar and late sericite. The quartz typically exhibits
strained extinction indicative of deposition within a stress regime. Fluid inclusion data
indicates that quartz has been (see following sections) deposited from either dilute circulating
meteoric-dominated fluids, or in some cases hypersaline magmatic-derived fluids.

3. The sulphide content of veins is interpreted to have been deposited from fluids with a
significant magmatic component, and is categorised as:

i) Fe-sulphide ± magnetite/hematite component, which is commonly massive, and

overgrows or infills fractures and breccia matrix. Pyrite is almost ubiquitous, and the
other phases are zoned from early and proximal to late and distal settings to the source
intrusion as: magnetite/hematite, through pyrrhotite, to arsenopyrite or arsenean pyrite.
The Fe-sulphides commonly contain inclusions of chalcopyrite and base metals.

ii) Chalcopyrite postdates the Fe-sulphide/oxide phases and decreases in abundance and
significance away from the source intrusion. Bornite is locally encountered in envi-
ronments transitional to a porphyry copper setting. At depth chalcopyrite infills frac-
tured in shattered and brecciated pyrite; whereas at shallow levels trace chalcopyrite
overgrows earlier phases, or occurs as inclusions in late auriferous pyrite. The copper
phases are typically accompanied by Bi-Ag-Pb-Te mineralization and locally W-Sn
phases, depending on the metal content of the source melt.

4. In some cases the sulphide event continues into, or is postdated by, a late carbonate (or as
at Lihir, anhydrite) event. The carbonate typically occurs as calcite and is inferred to be of
magmatic origin.

Gold mineralization in quartz-sulphide systems is typically associated with the sulphide event.
Refractory gold occurs within pyrite or As-rich pyrite distal to the source intrusion, possibly in
environments of rapid cooling. Non-refractory gold occurs as inclusions in coarser grained
pyrite or chalcopyrite formed under deeper conditions of slower cooling in environments more
proximal to the source intrusion. The latter commonly display a relationship to Bi-Ag tellu-
rides. Non-refractory gold in the porphyry^related quartz vein systems typically displays a
fineness range of 850-950, transitional between the fineness of gold in porphyry copper -gold
and carbonate-base metal gold systems (Fig. 4.8).

105
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

d) Examples

i) I.adolam gold deposit, Lihir Island, Papua New Guinea

The Ladolam gold deposit, Lihir Island (Fig. 1.2), has a mineable reserve of 14.6 M oz Au, and
contained gold of 42.6 million oz (Niugini Mining Annual Report, 1994). The Luise calde-ra was
targeted in 1982 by the Niugini Mining/Kennecott Joint Venture, which at the time was
evaluating the adjacent Tabar Island Group in Joint Venture with Nord Resources Limited. Pro-
specting quickly passed in the 1983 period from the identification of mineralized boulders along
the beach, to the delineation of soil geochemistry anomalies and drill testing Coastal (1983) and
Lienetz (1984) Zones (Davies and Ballantyne, 1987; Niugini Mining Annual Re-port, 1994). The
Minifie Zone was identified in late 1986 from a weak soil anomaly in a hand dug trench in an
area of scree on the caldera wall, and drill tested in the 1987 period.

During the Oligocene-Miocene the northward moving Australian plate was obliquely subducted
under the westward moving Pacific plate and the New Ireland and New Britain calc-alkaline
island arcs formed above a south dipping subduction zone (Fig. 1.2). The Pliocene collision of
the Otong Java Plateau jammed this subduction and a new north dipping arc formed south of
New Britain (Fig. 1.2). Most petrogenetic-tectonic analysis of the region favour models of re-
melting of oceanic crust to give rise to the shoshonitic volcanism which characterises the Ta-bar-
Lihir-Feni-Tanga island arc (Solomon, 1990; Solomon and Groves, 1994, McGinnis and
Cameron, 1994). North-south trending rifts in the overlying plate, inferred from seismic and
gravity anomalies and the alignment of volcanism, host mantle-derived material which consti-
tutes the Tabar-Lihir-Feni-Tanga island chain (Shatwell, 1987; Lindley, 1988; Marlow et al.,
1988; McGinnis and Cameron, 1994).

The Ladolam gold deposit, Lihir Island is hosted by the youngest of several overprinting vol-
canic edifices, possibly formed by the progressive unroofing by sideways collapse of a volcan-ic
edifice, now evident as the Luise Harbour (Fig. 7.4). The islands of the Tabar-Lihir-Feni-Tanga
arc were built up as Plio-Pleistocene shoshonitic volcanoes (Wallace et al., 1983).The structure
of the Ladolam gold deposit is dominated by north-south structures which reflect the
volcanoplutonism-hosting rift, ring fractures and possible conjugate fractures. Much of the cur-
rent thermal activity occurs along the ring fractures and the Minifie mineralization is localised at
the intersection of a throughgoing structure with the ring fractures (Fig. 7.4).

The Ladolam gold mineralization described as porphyry gold (Sillitoe, 1989), and displaying
epithermal gold characteristics (Moyle et al., 1990, 1991), while Carman (1994a, 1994b) em-
phasises the porphyry-epithermal telescoping. Subdivision of traditional (porphyry-related) ep-
ithermal ore systems (Leach and Corbett, 1995) allows the Ladolam mineralization to be cate-
gorized here as of the quartz-sulphide gold-style and also displaying alkaline porphyry affini-
ties. Note that the shoshonitic host rocks yield an alteration assemblage enriched in K-feldspar
(characteristic of this style) but depleted in quartz.
A sequence of three stages of overprinting alteration and mineralization is inferred from re-
views of the published literature (Davies and Ballantyne, 1987; Moyle et al., 1990, 1991;
Plimer et al., 1988; Forth, 1994; Carman, 1994b) and personal observations as:

Stage I: Host Intrusion Event

Emplacement of differentiated equigranular biotite monzonite, which grades from microdiorite
to syenite in composition, pre- and postdates andesite to latite porphyry stocks and dykes. The
intrusions have been emplaced into silica-poor alkali-basaltic (shoshonitic) volcanics, locally
forming contact breccias. Alteration related to emplacement of the intrusions is zoned from

106
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

potassic (biotite + K-feldspar) at depth to propylitic (epidote-chlorite-calcite + actinolite) at

shallow levels and peripheral to the intrusions. Biotite alteration has been dated at 0.9-1.0 m.y.
Late stage phyllic (sericite—illite) overprint is inferred to be related to the collapse of meteoric
waters onto the intrusion and associated breccias, and is possibly associated with very low grade
porphyry-style copper mineralization.

Stage II: Mesothermal Event

Later intrusion (Fig. 7.5) results in an overprinting sequence of events as: early brecciation, fol-
lowed by alteration and mineralization characterised by the deposition of sequential K-feldspar,
sulphides, and then anhydrite ± carbonate ± sulphides. Carman (1994b) describes a second po-
tassic alteration and breccias which appear to be indicative of phreatic and phreatomagmatic
(diatreme) eruptions (Section 3.ix.d.2). The latter are typical of environments characterised by
high level intrusions. This, locally polyphasal sequence of events is similar to that encountered in
most southwest Pacific rim porphyry-related low sulphidation systems as quartz-sulphide
mineralization. The host rock composition influences the alteration to K-feldspar-rich and
quartz-poor. The diatreme breccias have undergone subsequent extensive K-feldspar alteration
and associated leaching of plagioclase, sericite and mafics to produce a brittle, vughy, K-feldspar
rock. This leaching is especially intense in the breccias, and has formed a subhorizon-tal vughy
zone along the contact breccia at the top of the intrusive between Lienetz and Coastal zones,
which has previously been called 'the boiling zone' (Davies and Ballantyne, 1987). The vughy K-
feldspar rock is similar in appearance to the vughy residual silica/quartz zone in high
sulphidation systems. However, at Lihir the leaching is postulated to result from rock reaction by
neutral pH, silica-undersaturated fluids which are in equilibrium with the potassic and basic
volcanics. K-feldspar was also deposited in fractures and open vughs as either orthoclase or
adularia, and has been variably dated at 0.2-0.7 m.y. The younger ages are probably related to
later hydrothermal events.

The brittle K-feldspar altered volcanics and intrusions have undergone fracturing and associ-
ated sulphide deposition in veinlets, fluidised breccias, and sheeted veins, as well as lining or
infilling leached vughs. The sulphides are dominated by pyrite and arsenian pyrite, minor mar-
casite, and rare chalcopyrite, tennantite, sphalerite and galena. Sulphide mineralization contin-
ues into later fracturing and breccia events which are dominated by anhydrite ± carbonate dep-
osition. Quartz is absent at Lienetz and Coastal, but occurs as stockwork veins at Minifie. Bio-
tite, locally associated with the anhydrite-carbonate veins, has been dated at 0.34 my and re-
flects the high temperature of the porphyry-related-style of veins.

Native gold and gold-silver tellurides occur as submicroscopic inclusions in pyrite, mainly in
the sulphide phase, but extending into anhydrite-carbonate deposition. On the basis of fluid
inclusion, isotope and vein/alteration sequences, Plimer et al., (1988) relate gold mineralization
to the mixing of upwelling mineralized fluids, with cool meteoric waters. Gold mineralization
occurs at: Minifie as K-feldspar-pyrite alteration of breccias and later quartz stockwork veins
(Carman, 1994b), as pyrite coatings on the "boiling zone" breccias at Lienetz and Coastal zone,
where fluidised and crackle breccias are recognised at higher levels. Late stage quartz veins are
inferred to represent a cooler environment of quenching at the caldera margins.

Stage III: Geothermal Event

The current geothermal system is postulated to be related to the waning of the Stage II miner-
alization. Cristobalite-alunite alteration at shallow levels in previously formed breccia zones at
Lienetz and Coastal, grades laterally and at depth through kaolinite-silica to interlayered illite-
smectite alteration. This zoned advanced argillic—argillic alteration overprints earlier mineral-
ogy (Carman, 1994b), and is interpreted to be related to the draw-

107
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

down of low pH acid sulphate ± carbonate waters. These waters probably formed by oxidation
and condensation of gases which evolved from upwelling Stage II fluids. At <100°C, the low pH
waters could dissolve Stage II anhydrite and carbonate and deposit these phases in hotter
environments at depth. Alunite has been dated at 0.15 my., and is currently forming in the acid
sulphate springs at the surface. It is interpreted that waning of the hydrothermal system has
caused pressure draw-down of the cool acidic fluids, which are postulated to be remobilizing
Stage II copper-gold to form luzonite, enargite and locally coarse gold in the Stage III event.

ii) Kidston, eastern Australia

The Kidston gold deposit which contains >4 M oz Au, began production with a resource of 2.7
M oz Au at an average gold grade of 1.58 g/t Au (Baker and Tullemans, 1990), and added >1 M
oz at a grade of 1.25 g/t Au in a new ore zone in early 1995. It is hosted within a brec-cia pipe
related to Permocarboniferous volcanoplutonic activity which transects Precambrian basement
rocks. In the region between the Wirra Wirra and Lochaber volcanoplutonic com-plexes
(respectively NW and SW of Kidston), an arch of gravity contains outcropping Precam-brian
rocks which are intruded by quartz feldspar porphyry and rhyolite dykes, indicative of
underlying felsic Permocarboniferous intrusive rocks (Fig. 7.6). The Kidston breccia pipe is
inferred to have been derived from a magmatic source (Baker and Andrew, 1991) localised at
the intersection of a throughgoing structure, the Gilberton lineament, and the margin of the
buried arch of inferred Permocarboniferous intrusive. In much the same manner as individual
porphyry deposits, mineralized fluids may have migrated to the margin of the buried magma
source defined by the arch. The Gilberton Lineament is one of many parallel structural corri-
dors which display protracted histories of movement, including extension which facilitated the
emplacement of the Permocarboniferous volcanoplutonism (Fig 7.6; Laing, 1994; Corbett, un-
publ. data, 1983).

The Kidston breccia pipe is a teardrop shaped body about 1200 x 800 m, elongate in the trend of
the Gilberton Lineament. The breccia types reflect the transition from intrusive breccias with a
high proportion of strongly milled, introduced intrusive fragments, to peripheral co l-lapse-style
breccias which contain dominantly less milled host rock fragments (Fig. 7.7; Cor-bett, unpubl.
data, 1983). The metamorphic foliation in basement fragments is progressively reoriented as
fragments undergo increased milling and rotation away from the pipe margins. The continuation
of a geological contact in the basement rocks as differing breccia types is in-dicative of the
collapse nature of the peripheral breccias (Fig. 7.7). Baker and Andrew (1991) delineate a model
of overprinting brecciation in which features such as: clasts of quartz-pyrite-magnetite
stockwork veining, tourmaline in fragments and breccia matrix, dykes which in-trude the
breccia, and isotope data, are indicative of an intrusive source at depth. These work-ers describe
insitu exfoliation as a mechanism for the rounded intrusive fragments cited above, and suggest
that the pipe did not breach the surface. Rhyolite dykes tend to predate pipe for-mation, and later
quartz feldspar porphyry dykes which transect the breccia, predominantly in the central portion
of the pipe, may represent a tapping of a deeper portion of the magma source (Fig. 7.7, Corbett,
unpubl. map, 1980; Baker and Tullemans, 1990). Sheeted fractures which form kinked
polygonal shapes about the pipe margin cross cut the breccia as a final phase of post-brecciation
collapse (Fig. 7.7).

The style of mineralization is evident from detailed work on the Kidston breccias (Baker, 1987;
Baker and Tullemans, 1990; Baker and Andrew, 1991). The following three main stages of
alteration, veining/brecciation and mineralization have been recognised by these authors at
Kidston (Fig. 7.8):

108
Fig. 7.6

Fig. 7.7
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure. Alteration, and Mineralization 1 Corbett G J & Leach T M. 8/96 Edn.

Stage I: Rhyolite Intrusion:

Emplacement of a high level rhyolite stock was accompanied by quartz stockwork veins
development and minor molybdenite-pyrite ± arsenopyrite ± chalcopyrite mineralization.
Tourmaline-sericite ± andalusite breccias formed at the margin of the rhyolite stock at Wise's
Hill during late stage volatile exsolution. Fluid inclusion data is indicative of hot (400-500°C)
and hypersaline (>40-50 wt percent NaCl) conditions during Stage I activity at depths of 3 km.

Stage II: Syn- and Post-Breccia Pipe Event:

The breccia matrix at Kidston has undergone alteration to: an early potassic assemblage of biotite
± muscovite ± epidote alteration and pyrite-pyrrhotite mineralization, followed by later
muscovite-epidote-orthoclase-calcite alteration (propylitic) and pyrite-pyrrhotite ± sphalerite
mineralization. Minor gold mineralization is associated with epidote in the latter phase of activ-
ity. Fluid inclusion data indicates that moderately high temperatures (260-360°C) and salinities
(2-10 wt percent NaCl) prevailed.

Stage III: Quartz-Carbonate Veins:

This is the main event of gold mineralization and occurs within sheeted veins around the mar-
gins of the breccia pipe (Figs. 7.7, 7.9) and is interpreted to have been sourced from the same
melt as the post-breccia quartz feldspar dykes.

The sheeted veins consist of subparallel fractures, up to 10 cm wide, which exhibit a deposi-
tional sequence of quartz + sericite —> sulphides —> carbonates, typical of quartz-sulphide and
carbonate-base metal gold systems elsewhere in the southwest Pacific. Lateral and vertical
zonations are recognised in both gangue and ore phases (Baker, 1987; Leach, unpubl. data).
Carbonate (mainly ankerite) dominates over quartz at shallow levels, whereas quartz is domi-
nant over carbonate (calcite) depth and marginal to the ore zones at Wises Hill and North Knob.
The iron phases are zoned pyrite —> arsenopyrite —> pyrrhotite ± magnetite with in-creasing
depth and laterally away from the ore zones. Sphalerite and galena mineralization is restricted to
shallow levels in the ore zones, whereas chalcopyrite persists to depths.

Gold mineralization preferentially occurs in the shallow quartz-ankerite/calcite-pyrite and ar-

senopyrite zones as inclusions in sphalerite and pyrite, infilling fractures and overgrowing py-
rite and marcasite, and as intergrowths with Bi- and Ag-tellurides and galena. High fineness
gold (906-911) is associated with quartz-sulphides-Bi-phases, moderately high fineness with
carbonate-base metals (768-891), whereas low fineness gold (<500) infills late fractures.

Fluid inclusion and mineral isotope data (Baker and Andrew, 1991) indicate that Stage III
quartz and carbonate were derived from different fluids. Quartz was deposited over a broad
temperature range (120-300°C) and at relatively saline (2-10 wt percent NaCl) conditions,
whereas carbonates were deposited over a similar temperature range but from significantly
more dilute (0.2-0.7 wt percent NaCl) waters. Isotope data shows that the carbonates were
precipitated from fluids containing a significantly higher meteoric component than the quartz.
Two-phase fluid inclusions, indicative of boiling, are absent.

Magmatic-derived mineralized fluids are inferred to have migrated up along sheeted fractures in
the Wise's Hill and North Knob areas. Increased fluid flow is anticipated at the intersections of
sheeted fractures and those fractures dilated by the extensional tectonism to which the vol-
canoplutonism is related. Sulphide and gold mineralization in the sheeted veins is therefore pos-
tulated to have taken place in response to the mixing of these upwelling saline and

109
Stages of Alteration, Veining and Mineralization at Kidston (data from Baker & Andrew, 1991)

Fig. 7.8

Fig. 7.9
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

magmatic-derived fluids which deposited quartz, with dilute meteoric waters which deposited
carbonate (Fig. 7.9). The current level of erosion of the sheeted veins has exposed a system
which is transitional from a carbonate-base metal gold style mineralization at shallow levels, to
a quartz-sulphide-style mineralization at deeper levels and in zones peripheral to inferred fluid
upflow zones.

iii) Bilimoia District, Papua New Guinea

Porphyry intrusion and mesothermal gold bearing veins occur north of Bilimoia village (Fig.
1.2), near Kainantu in the Eastern Highlands of Papua New Guinea (Corbett et al., 1994b). Host
rocks comprise Early Mesozoic Bena Bena Formation phyllites which are intruded by
syntectonic Karmantina Granite Gneiss and Mid Miocene Akuna Granodiorite (Rogerson et al.,
1987). Porphyry copper-gold mineralization elsewhere in the region is associated with Late
Miocene Elendora Porphyry intrusions (Rogerson and Williamson, 1985). This area is situated
immediately south of the Markham Fault, the suture between the Pacific and Australi-an plates
(Fig. 1.2). Two inferred porphyry centres are localised by the intersection of transfer structures
with structures formed parallel to the New Guinea Orogen (Fig. 7.10). Mesothermal veins are
hosted within pre-mineral, arc-parallel, structures at Bilimoia, and arc-normal struc-tures at
Arakompa (Corbett, 1994; Corbett et al., 1994b).

At Bilimoia the slaty cleavage within the Bena Bena schist basement rocks varies to a crenu-
lation cleavage within the arc parallel structures. As such a foliation must have formed at depths
in the vicinity of 5 km (D. Grey, pers. commun., 1992), these represent major arc par-allel
structures which have been reactivated and mineralized at higher crustal levels. Minerali-zation
extends for strike lengths of up to 2 km along a series of sub parallel structures which occur as:
slickensided silicified fault faces, puggy shears, and are locally exploited by dykes or fluidised
breccias (Fig. 7.10).

Mineralization is exposed over a vertical extent of over 800 metres on the steep slopes be-
tween the Markham Valley and Eastern Highlands. Generally NS trending, higher grade ore
shoots occur within or adjacent to the NW trending mineralizing structures, commonly at the
intersections of cross structures (Corbett et al., 1994b). The local sigmoidal shapes are con-
sistent with the inference that these ore zones formed as tension gash features during dextral
rotation on the controlling structures (Corbett, 1994).

The following three stages of vein and breccia development have been identified at Bilimoia
(Fig. 7.11; Corbett et al., 1994b):

1. Shearing and Brecciation:

An initial stage of shearing and fracturing is locally accompanied by fine fluidised breccias
composed of quartz-sericite-pyrite, and coarser diatreme breccia-like intrusive fluidised brec-
cias containing dyke fragments which are correlated with the Elandora porphyry of Rogerson
and Williamson (1985).

2. Quartz Veins:
A major event of extensive fracturing and brecciation, formed locally polyphasal crackle
breccias or open veins, consisting of clear to milky quartz, grading outwards to crustiform or
coxcomb quartz which extends into open cavities. The quartz commonly exhibits strained ex-
tinction indicative of deposition within a stress regime and in places is accompanied by minor
sericite, pyrite and yellow sphalerite. Wall rock phyllites have locally undergone alteration to
the Cr-rich micas, mariposite and fuschite. Fluid inclusion data indicates that the quartz veins
were deposited from dilute (>2 wt percent NaCl, very locally >3-4 wt percent

110
 Fig. 7.10

Paragentic Sequence of Veining and Mineralization at Bilimoia

Fig. 7.11
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

NaCl) waters over a wide temperature range of 210-330°C. There is no consistent zonation in
temperatures or salinities over the 700 metre vertical extent of veins. It is therefore interpreted
that quartz deposition took place from dilute meteoric waters circulating within deep crustal
structures. In places the quartz is chalcedonic, radiating, or fibrous, indicative of rapid quench-
ing conditions. The presence of local interlayered illite-smectite as a wall rock alteration also
implies periodic recharge of cool fluids.

3. Pyrite ± Base Metals:

Quartz vein development is followed by deposition of massive to coarse grained pyrite + fine
quartz in open cavities and fractures, or in thin fractures cutting the metasediment wall rock.
The local intergrowth of pyrite with base metal sulphides (sphalerite, galena, chalcopyrite and
tennantite), trace magnetite, and common minute inclusions of chalcopyrite, bornite, and hy-
pogene covellite, are indicative of the development of these veins as a precursor to the main
copper-gold mineralization which followed.

4. Copper Mineralization:
Chalcopyrite overgrows pyrite, and in places infills fractured and shattered pyrite with associ-
ated fine grained quartz-sericite deposition, and locally forms intricate intergrowths with born-
ite. At the Karempa (Fig. 7.10), pyrite-chalcopyrite mineralization is accompanied by deposi-
tion of topaz-sericite, diaspore-dickite or sulphates (anhydrite, barite). This is indicative of a
periodic influx of moderately low pH, sulphate-rich magmatic dominated fluids.

A wide range of mineral phases characterised by W-Sn, Bi-Te-Ag and Cu-As-Sb minerali-
zation accompanies chalcopyrite deposition. These phases are also indicative of the inclusion
of late stage fluids with a significant magmatic component, probably derived from emplace-
ment of an Elandora-style silicic felsic porphyry intrusion at depth. The paragenetic sequence
of mineralization:

Te --------------- > Pb,Ag,Bi ----------------- > Sn - W --------------- > Cu, As, Sb

is consistent with decreasing fTe2 and increasing fS and xCu with time.

The initial deposition of tellurium-rich phases occurs as: native tellurium which is overgrown
by tellurobismuthinite (Bi2Te3), followed by lead (altaite - PbTe), and silver (Ag2Te) tellurides.
Later bismuth-rich phases include tetradymite (Bi2Te2S), bismuthinite (Bi2S3) and Bi-rich gale-
na. Tin and tungsten phases appear to post date Bi-Ag-Te mineralization. The Fe-wolframite
phase ferberite, is relatively common and is overgrown by Sn-Cu phases such as mawsonite
(Sn-rich bornite), and a Sn-As-covellite species. Local Cu-Bi-Te sulphides such as aikinite
(Cu[Pb,Bi]2S3), goldfieldite (Cu[Te,Sb]S4) and Bi-rich enargite are interpreted to be transitional
between the early bismuth-telluride, and late copper phases of mineralization, characterised by
chalcopyrite and minor bornite. The infilling of fractures in massive pyrite by native cop-per,
and the formation of covellite and chalcocite, are interpreted represent supergene alteration
phases of primary chalcopyrite.

The supergene gold, in oxidised quartz veins mined by the local villagers, commonly exhibits a
"mustard" texture, indicative of a primary source association with telluride phases. At depth, the
gold occurs as inclusions in chalcopyrite, and as inclusions within, and overgrowing tellu-rides,
bismuthinite, and hessite incorporated within the chalcopyrite. Near the Kora mine (Fig. 7.10),
gold is encountered as inclusions in ferberite, and associated pyrite. Primary gold has a fineness
of 834-922 (average 858), which is characteristic of quartz-sulphide vein systems formed
peripheral to porphyry intrusions elsewhere in the Pacific region (Fig. 4.8). Gold in the Yar Tree
Hill prospect, 7-8 km along strike southeast of the Bilimoia quartz vein

111
 Fig. 7.12

Interpreted Paragenetic Sequence of Arakompa Veining and Mineralization

Fig. 7.13
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

systems, is also encountered as inclusions in chalcopyrite cutting pyrite, and has a similar
fineness (860-940, average 895) to gold at Bilimoia.

It is interpreted that early stage quartz veins and wall rock sericite-fushcite alteration were de-
posited in response to cooling of dilute meteoric water-dominated fluids which circulated along
the deep crustal Bilimoia structures. The chromium (in fuchsite) appears to have been derived
from the migration of these fluids through ultramafic host-rocks at depth (ultramafics outcrop to
the north-west of the Kainantu region). The emplacement of Elandora porphyries along these
structures has resulted in initial intrusion of fluidised breccias, and shallow level felsic dykes as
the precursors to the introduction of mineralizing fluids, which resulted in the deposition of
chalcopyrite-pyrite-gold and associated Bi-Te-W-Sn-Ag mineral phases.

Zonations in styles of alteration, veins and mineralization at Bilimoia provide vectors which
point towards an inferred buried intrusive source for the gold mineralization (Fig. 7.12), in the
vicinity of a landslip in phyllic alteration (Fig. 7.10). Potassic alteration hosted in Akuna
granodiorite and containing weak copper mineralization at Kokofimpa, is overprinted by phyl-lic
alteration extending SE to Bilimoia village. Magmatic volatiles which evolved to the south and
west from the buried porphyry resulted in the formation of the extensive and pervasive shoulder
of advanced argillic (high sulphidation) alteration (Fig. 7.10). This is locally transect-ed by
structurally controlled enargite mineralization at the Headwaters Prospect (Fig. 7.10).
Mineralized fluids migrated along NS structures and laterally along pre-existing NW-SE
structures to form mesothermal-style mineralization which displays a progressive zonation as:
Cu ± Au, Au-Cu and Pb-Zn, at increasing distances from the inferred porphyry source (Fig.
7.12). Higher gold grade ore-shoots formed within localised dilational jogs and at sites of
quenching localised by cross structures.

At Arakompa, mesothermal quartz veins occur proximal to sub-economic porphyry copper-gold

mineralization (Fig. 7.10; Corbett, 1994; Corbett et al., 1994b). Host rocks are the mid Miocene
basement Akuna Granodiorite, and mineralization may be related to younger Elando-ra-style
porphyry intrusives which crop out in the area (Rogersbn and Williamson, 1985). Pre-mineral
NNE trending arc normal structures have undergone dilation during subduction-related
compression and so correspond to the tensional vein setting discussed in Section 3.viii.
Intersections with NE trending arc parallel structures represent sites of stockwork veining. Lo-
cal jogs in the controlling faults localise thicker lodes, commonly with elevated gold and cop-per
grades. Most mineralization is confined to fault -controlled gossanous lodes and adjacent pug
zones which display some post mineral movement. Both rock types are worked by local miners
for supergene gold. Pebble dykes recognised in drill core are indicative of pre -mineral explosive
magmatic fluid emplacement along the fault structures, and are cut by quartz and sulphide veins.
Some contain exotic basement rock fragments carried up from unknown depths.

Four stages of veining have been categorised at Arakompa (Figs. 7.13, 7.14; Corbett et al.,
1994b) as:

1. Pebble Breccia Dykes:

The Arakompa structures contain breccias comprising well milled fragments of: phyllic altered
Akuna granodiorite, hornfelsed sediments, and quite low metamorphic grade phyllites similar
to those which crop out at Irumafimpa, and rare early quartz-sericite-pyrite vein clasts.

1. Quartz Veins:
Extensive coarse-grained, cockscomb to locally banded quartz deposition, is accompanied by
coarse cubic pyrite, sericite, as well as local epidote, magnetite, and carbonate, and are locally

112
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure. Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

cut by pebble dykes. Fluid inclusion data indicates that the quartz was deposited over a wide
temperature range (245-315°C, average 285°C), under local two-phase (boiling) but dilute (< 2
wt percent NaCl) fluid conditions, similar to quartz veins at Bilimoia.

3. Polyphasal Fracturing and Brecciation:

Brecciation of the early quartz veins and pebble breccias is accompanied by deposition of mas-
sive fine to coarse grained pyrite, and fine granular quartz. The pyrite commonly contains inclu-
sions of chalcopyrite, bornite, sphalerite, galena and rutile, indicating that this event is a precur-
sor to the Stage 4 mineralizing event. Limited fluid inclusion data on Stage 3 quartz suggests
that the brecciation and quartz-sulphide veining took place in response to an influx of fluid at a
similar temperature (250-290°C), but of significantly higher salinity (4-6.5 wt percent NaCl),
than the earlier quartz veins (Fig. 7.14).

4. Copper-Gold Mineralization:
Chalcopyrite and minor quartz-sericite overgrow the earlier mineral phases. Local fracturing
and in situ brecciation has accompanied the copper mineralization. The local deposition of
carbonate with chalcopyrite, is indicative of a gradation to a carbonate-base metal style of
veining, present at Maniape (Fig. 7.10). A wide array of Bi-Ag-Pb-Cu ± Zn ± Sn telluride and
sulphide phases (hessite, tetradymite, bismuthinite, cuporparonite, witticherite and ham-merite)
were deposited either transitional between Stage 3 pyrite and Stage 4 chalcopyrite, or
contemporaneous with the chalcopyrite mineralization (Corbett et al., 1994b). Trace tin phases
(stannoidite, kesterite and a Te-cannfieldite) are also associated with chalcopyrite deposition.

Gold at Arakompa occurs as native Au°, generally as inclusions in Stage 3 pyrite and Stage 4
chalcopyrite, and is commonly associated with Ag-Bi-Cu-Pb-Te ± Sn/Zn phases. Gold dis-plays
a high fineness (723-995, average 877), characteristic of quartz-sulphide gold systems,
transitional between porphyry copper-gold and carbonate-base metal gold systems (Fig. 4.8;
Leach and Corbett, 1994, 1995). The higher salinity fluids and presence of Bi-Te phases dur-
ing Stage 3/4 activity is indicative of a significant influx of magmatic-derived fluids during the
formation of mineralization at Arakompa.

The increase in copper contents and fluid inclusion temperatures with depth, further suggests
that the mineralized fluids have migrated from a porphyry in the vicinity, and probably below
the Arakompa Prospect. These fluids have moved upwards depositing copper and gold into re-
opened quartz and pyrite-quartz veins at Arakompa, as well as south and west to form the more
distal, and dilated, Maniape pre-existing, quartz-pyrite vein structures (see below). A prominent
aeromagnetic high at Arakompa may be related to a magnetite-bearing potassic al-tered
porphyry at depth. Weak copper-gold mineralization is associated with outcropping po-tassic
alteration at nearby Nontifa (Fig. 7.10).

The Maniape prospect is discussed as a carbonate-base metal style of gold mineralization in

Section 7.iii.j.v.

iv) Hamata, Morobe Goldfield, Papua New Guinea

At Hamata, dipping veins and shears occur in the hanging wall of the Upper Watut Graben
Fault (Fig 7.28). The structural and geological setting of alteration and mineralization at Ha-
mata are considered more fully in the discussion of the Morobe goldfield (Section 7.iii.j).

Gold mineralization at Hamata is hosted in Morobe granodiorite and occurs in at least two
subparallel zones (Masi and Lower Zone, Denwer et al., 1995; Wells and Young, 1991). The

113
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbetl G J & Leach T M, 8/96 Edn.

Morobe granodiorite has undergone intense K-feldspar - sericite alteration within these zones,
which overprints high grade propylitic (actinolite-epidote) and local potassic (biotite) altera-
tion. Veins within these zones, which are up to 50 m thick, is diffuse except where 3-4 m wide
'reefs' of pyrite-hematite-magnetite-quartz veins are well developed.

The paragenetic sequence of veining and mineralization at Hamata may be summarised from
Denwer et al., 1995 as:

1. Early thin veinlets of magnetite, hematite and pyrite exhibit K-feldspar selvages, and are
overgrown and cut by K-feldspar - quartz veins. The quartz characteristically contains two
phase inclusions and halite daughter phases. Fluid inclusion data indicates that the quartz was
deposited under relatively hot (270-340°C) and periodic hypersaline (up to 35 wt percent
NaCl) to moderately saline (3-7 wt percent NaCl) conditions, indicative of an environment
proximal to a porphyry system.

2. A major stage of massive sulphide-oxide vein development was accompanied by fine

grained quartz-sericite deposition. Specular hematite and coarse lath-like magnetite overgrow
early pyrite. Chalcopyrite locally seals shattered pyrite, and infills fractures with sericite-pyrite
cutting early quartz. Bi-Te (tetradymite) and W (ferberite) mineralization is associated with
pyrite-chalcopyrite deposition. Native gold infills fractures and cavities in pyrite, is closely
associated with Bi-tellurides and has a fineness of 816-991 (average 911). The style of gold
mineralization at Hamata is therefore very similar to that encountered at Arakompa and
Bilimoia.

3. The deposition of local carbonate-base metal sulphide veins in which pyrite, calcite and
chalcopyrite are the dominant phases, is accompanied by minor sphalerite, galena and late
stage arsenopyrite. In places hematite and magnetite deposition locally extends into this
carbonate-base metal phase of alteration.

4. Late stage quartz and/or barite veins contain local arsenopyrite.

The Hamata deposit occurs at a much lower elevation than the other deposits in the Bulolo
Graben and may therefore be considered to represent a quartz-sulphide vein system transi-
tional between the carbonate-base metal gold systems (higher) and an inferred buried porphyry
copper-gold source for the alteration and mineralization. The Hamata deposit crops out along
the strike of the same structure as the Hidden Valley carbonate-base metal gold de-posit, but at
a several hundred metre lower elevation, in keeping with the overall zoneation of these deposit
types.

v) Exciban, Philippines

Gold mineralization in the Exciban deposits, Camarines Norte district, Philippines, displays
characteristics typical of porphyry-related quartz-sulphide vein systems. The following dis-
cussion is taken from James and Fuchs (1990) and Mitchell and Leach (1991).

Mineralization occurs within a set of steeply dipping NNE-trending structures formed at a high
angle to the Larap thrust zone. Early quartz veining is hosted in weakly metamorphosed and
locally sheared volcanics and arenaceous sediments. The quartz contains abundant halite
daughter crystals indicative of hypersaline fluid conditions. Massive sulphide veining is char-
acterised by pyrite and chalcopyrite, the later commonly as overgrowths and infilling of frac-
tures and cavities in the pyrite. Native gold occurs as inclusions in pyrite and chalcopyrite and
is generally associated with bismuth telluride phases (tellurobismuthinite, tetradymite and

114
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

hedleyite [Bi7Te3]).
James and Fuchs (1990) infer a magmatic-dominated source for the veins and mineralization
based on the presence of daughter crystals in fluid inclusions, abundance of telluride phases and
high copper content of the veins. Dacite dykes crop out at the surface and are interpreted to
represent high level equivalents of the mineralizing porphyry at depth. These workers attribut-ed
high cobalt levels in the veins to have been derived from the mafic/ultramafic host rocks at
depth.

iii) Carbonate-Base Metal Gold Systems

a) Introduction

A class of porphyry-related gold mineralization, associated with carbonate-base metal veining

and breccia infill, forms at intermediate levels between southwest Pacific rim porphyry and ep-
ithermal environments (Figs. 7.1, 7.2; Leach and Corbett, 1993, 1994, 1995). Sillitoe (1989) and
Handley and Bradshaw (1986) alluded to the existence of this class of deposits in emphasising
the magmatic association and noting an overlap between the epithermal and porphyry envi-
ronments, especially in relation to the Porgera gold deposit. Most deposits within this class have
until now been categorised as adularia-sericite epithermal gold-silver, and yet many lack adularia
which, where present, is not related to the gold mineralizing events. In addition, many form at
levels transitional between epithermal and porphyry environments. This more detailed
classification is possible following the upsurge in gold exploration during the 1980's.

Classic low sulphidation adularia-sericite epithermal gold-silver deposits, (e.g., Hishikari and
Sado, Japan; Waihi and Golden Cross, New Zealand), are dominated by quartz and adularia
within fissure veins. However, much of the gold is not associated with quartz-adularia but oc-
curs in sulphide bands, (ginguro ore in the Japanese literature). Some systems characterised by
chlorite, (e.g.. Cracow, eastern Australia), or illite (e.g., Tolukuma, PNG), are inferred to have
developed from ore fluids which display typical epithermal meteoric as well as magmatic char-
acteristics. Although carbonate-base metal gold deposits may also form as fissure veins, the
associations with base metals high level porphyry intrusions are indicative of a transitional set-
ting between the epithermal and porphyry environments. Comparison between many deposits in
the southwest Pacific rim allows the carbonate-base metal gold deposits to form a class of their
own (J^each and Corbett, 1993, 1994, 1995). An understanding of the anatomy and fluid flow
paths from the alteration zonation and structure, may point towards high gold grade portions of
carbonate-base metal gold systems, or the porphyry source rocks.

b) Definition

Carbonate-base metal gold systems develop distal to porphyry intrusives from the mixing of a
magmatic derived fluid with surficial bicarbonate gas condensate waters (Figs. 1.4, 2.4). Min-
eralization varies from higher grade vein/breccia lode mineralization to bulk low grade frac-ture
or breccia infill styles. Major structures localise hydrothermal systems, and by movement create
dilational ore-hosting environments in subsidiary structures. High level porphyry intru-sions are
commonly spatially associated with ores and may represent competent host rocks. Maar
volcano/diatreme breccia complexes and intrusive fluidised breccias, occur as pre-mineral
phreatomagmatic explosive events which focus fluids degassing from porphyry bodies at depth,
and also create fracture permeability as ore-hosting environments.

115
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure. Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Base metal contents typically occur as Zn > Pb > Cu, while carbonates exhibit a wide range in
chemistry and spatial zonations from Fe-, to Mn-, Mg-, and Ca-carbonates, with increasing depth
(Fig. 7.17). Gold mineralization preferentially occurs in association with the Mn/Mg car-bonates.
There is a progression in time and space (crustal level) from porphyry to epithermal
environments. Carbonate-base gold mineralization is commonly preceded by mesothermal to
epithermal quartz-sulphide veining, and locally porphyry-related quartz stockwork veining, de-
pending upon the depth of the system. Mineralizing fluids are transitional between dilute circu-
lating meteoric waters, typical of epithermal environments, and high temperature saline porphyry
systems.

c) Distribution

Some significant southwest Pacific rim carbonate-base metal gold systems are: in Indonesia,
Kelian (5.7 M oz contained gold, van Leeuwen, 1994), Busang (22.5 M oz Au), parts of Mt
Muro (1 M oz Au), and Cikotok (> 2 M oz Au); in Papua New Guinea, Porgera mineralization
types A, B and E (>6 M oz Au), Mt. Kare, the Morobe Goldfield group of deposits (past pro-
duction with alluvial 3.7 M oz Au, Lowenstein 1982), including Upper Ridges, Golden Ridges
and Golden Peaks at Wau, Edie Creek, Kerimenge (1.8 M oz Au, Hutton et al., 1990), Hidden
Valley (2.4 M oz Au, Nelson et al., 1990), Busai and Kulumadau on Woodlark Island (Corbett et
al. 1994a), and Maniape at Kainantu (Corbett et al., 1994b); in the Solomon Islands, Gold Ridge;
in the Philippines, Acupan (4 M oz Au; Mitchell and Leach, 1991); in eastern Australia, Mt.
Terrible (Teale, 1995), and Copper Hill (Leach unpubl. data). Some epithermal gold-silver
quartz-adularia-sericite deposits are now recognised to exhibit affinities with carbonate-base
metal systems (e.g., Tolukuma, PNG; Cracow, eastern Australia). Carbonate-base metal gold
mineralization at the Acupan (4 M oz Au) and Antamok (10 M oz Au), Baguio District, Philip-
pines, overprint uplifted porphyry copper-gold style mineralisation (Mitchell and Leach, 1990).

d) Geological setting

Carbonate-base metal hydrothermal systems form at elevated crustal levels above porphyry
copper-gold deposits, and so tend to be associated with higher level, possibly differentiated,
porphyry intrusions. Thus, the accretionary prism of moderately eroded island arc terrains is a
primary setting for these deposits, especially where competent metamorphic basement rocks
fracture to host fracture/vein systems. Intra arc rifts such as the Bulolo Graben (Fig. 7.28;
Corbett, 1994), may represent a locus of high level porphyry intrusion, resulting from crustal
thinning. Other intrusion centres such as Porgera (Corbett et al., 1995); Mt Kare (Corbett,
1994); Kelian (van Leeuwen et al., 1990); and Kulumadau (Corbett et al., 1994a), are localised
by major structures. Kelian, Busang and Mt Muro all occur on a major crustal suture (Fig. 1.2)
which separates rocks of different ages and markes the edge of the magmatic arc defiend by
Mitchell and Carlile (1994).

Many carbonate-base metal gold systems are associated with milled matrix fluidised hydro-
thermal (diatreme) breccias (e.g., Kelian, Sillitoe, 1994; Acupan, Domasco and Guzman, 1977;
Kerimenge, Akiro, 1986; Corbett, unpubl. report; Wau, Sillitoe et al., 1984; Woodlark Island,
Corbett et al., 1994a; Edie Creek, Corbett, 1994). Phreatomagmatic explosions which form
maar volcanoes/diatreme breccias result from the sudden heating of ground waters in contact
with a porphyry heat source (Section 3.viii.c.2). Pre-mineral intrusive breccias characterised by
milled, fluidised or muddy matrix may exploit pre-existing structures and prepare the plumbing
systems subsequently utilised by subsequent mineralized fluids. Major structures commonly
host ground waters, and are utilised by rising intrusives, and so may represent the locus of
breccia formation. Phreatomagmatic eruptions tap the top of the magma

116
 Exploration Workshop ■Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

chamber, which subsequently degasses to evolve off the mineralized fluids, and provide
fracture ground preparation of the adjacent competent host-rocks.

e) Structure

As discussed in the examples below, mineral deposition in carbonate-base metal gold systems is
promoted by the mixing of rising magmatic fluids with groundwaters in: fissure veins at di-
atreme margins (e.g., Kerimenge, Edie Creek, PNG; Acupan, Philippines), structures which act
as higher grade feeder structures to intervening tension gash vein/breccias which host lower
grade ore (e.g., Busai, Woodlark Island, PNG; Antamok, Philippines), to fractured and brecci-
ated intrusive margins (e.g., Kelian, Indonesia; Porgera, PNG), or dilatancy associated with
throughgoing strike-slip structures (e.g., Maniape, PNG). While bonanza gold grades are com-
mon in lode mineralization, elevated gold grades may also form by repeated deposition in dila-
tional structures (e.g., Acupan, Philippines; Upper Ridges, PNG), especially if proximal to fluid
upflow zones. Hanging wall splits are ideal settings for fluid mixing and hence mineral deposi-
tion (e.g., Kerimenge, Hidden Valley, Upper Ridges in the Wau district PNG).

Pre-mineral structures control fluid flow and fracturing about the margins of breccia bodies,
such as maar volcano/diatreme breccias, which then represent ideal loci for fluid flow and
hence mineralization. Thus an ideal setting for carbonate-base metal gold mineralization might
be fracturing near the intersection of major through-going structures and diatreme/maar com-
plexes (e.g., Upper Ridges, Kerimenge), or at the contact of veins with diatreme pipe margins
(e.g., G.W. breccia pipes at Acupan, Damasco and Guzman 1977). Fluidised breccias prepare
pre-existing structures exploited by carbonate-base metal veins at Busai, Woodlark Island
(Corbett et al., 1994a).

Rock competency is a critical factor in fracture development and hence mineral deposition. At
Porgera, intrusive stocks are inferred from the aeromagnetic data (Henry, 1988) to cap a much
larger intrusive system at depth (Corbett et al., 1995). The host Chim Formation shales are ex-
tremely incompetent and locally water-bearing. Degassing fluids from depth are preferentially
focused into the fractured margins of high level stocks and competent contact baked sediments.
The structural tapping of degassing fluids is most evident in the relationship of the later ros-
coelite mineralization to the Roamane Fault, a bounding structure to the intrusive complex.
Similarly, at Kelian, brecciated intrusive margins host mineralization, whereas the incompetent
'muddy breccias' are less well mineralized (van Leeuwen et al., 1990).

Maar volcano/diatreme breccias are gas driven and so display clay alteration of the breccias, and
do not fracture well in the upper portions which are characterised by lower temperature clays.
Thus, the fractured margins of pipe-like breccia bodies are the locus of fluid flow. Only in
deeper portions of diatreme breccias, where higher temperature clay alteration is more com-
petent, do diatreme breccia complexes host gold mineralization (e.g., Montana Tunnels, Sillitoe
et al., 1985). The suggestion by some workers (Sillitoe, 1989) that the host rocks to car-bonate-
base metal gold mineralization at Gold Ridge are diatreme breccias, may account for the poor
development of fracturing.

f) Alteration and mineralization

A sequence of overprinting events catagorised (Fig 7.15) for carbonate-base metal gold
systems is described as:

Stage 1 fluidised breccias which are interpreted to form in association with initial porphyry

117
Paragenetic Sequence of Veining and Mineralization in Carbonate - Base Metal Sulphide Gold Deposits.
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

emplacement at depth and localised propylitic alteration of the host rocks. Breccias commonly
occur as diatreme breccias (e.g., Wau, Kelian, Acupan, Kerimenge) or intrusive equivalents as
fluidised equivalents (e.g., Woodlark Island). Phreatomagmatic breccias provide pre-mineral
ground preparation and focus mineralization by tapping the top of the magma chamber.

Stage 2 veining is dominated by quartz and post-dates the breccia event, and in some deposits
contains either early adularia and/or late sericite/illitic clay. The quartz veining ranges from:
porphyry-related quartz stockwork (e.g., Copper Hill, eastern Australia, Leach unpubl. data;
Porgera, Richards, 1992) at deep levels, to quartz-sulphide veining at intermediate levels (e.g.,
Kelian, Indonesia; Morobe Goldfield, Woodlark, Maniape all in PNG; see examples below), and
to crustiform banded quartz-adularia at shallow levels (e.g., Tolukuma, PNG, Corbett et al.,
1994c). Fluid inclusion data indicate that the early quartz veining in many deposits was de-
posited from relatively hot (250-350°C), but dilute (<2-4 wt percent NaCl) fluids (Fig. 7.16).

Stage 3 carbonate-base metal sulphide veining typically overgrows and/or crosscuts the earlier
quartz veining. The base metal sulphides commonly predate carbonate veining and comprise:
early pyrite, followed by sphalerite and galena, overgrown by later copper phases (mainly
chalcopyrite followed by tennantite). In some instances, base metal sulphide mineralization ex-
tends into the carbonate event.

Stage 4 carbonates may occur as fine crustiform banded veins, locally alternating with thin
quartz-rich carbonate bands. Fluid inclusion data indicates that the carbonate and sphalerite in
many systems were deposited from a significantly saline fluid (<4-6 wt percent NaCl; up to
>20 wt percent in sphalerite at Kelian), but at similar or slightly lower temperatures, to the
earlier quartz. High salinity inclusions in carbonate and sphalerite are interpreted to indicate an
influx of a fluid with a significant magmatic component during the Stage 2-3 activity.

Gold mineralization predominantly develops during base metal sulphide deposition and extends
into the carbonate event. Gold typically occurs in the native state, either as inclusions in pyrite or
base metal sulphides, intergrown with carbonate, or infilling fractures and vughs in earlier
quartz. Some gold mineralization also occurs within late stage quartz veining, especially where
abundant pyrite/arsenopyrite is present (e.g., Kerimenge, Syka and Bloom, 1990; Porgera,
Richards, 1992). The average fineness of the gold in carbonate-base metal systems typically lies
within the range of 700-850 (Fig. 4.8), intermediate between epithermal quartz gold-silver
systems and quartz-sulphide gold veining formed marginal to porphyry intrusions.

Late stage activity is either dominated by surficial fluids which result in the deposition of
kaolin, interlayered clay, gypsum, and quartz; or deep fluids characterised by calcite dep-
osition. Gold mineralization may locally persist into the initial stages of this late stage
event.

g) Zonations in vein style and mineralization

Carbonate-base metal gold deposits exhibit distinctive zonations in styles of veining and min-
eralization, from regions proximal to a porphyry source or hot conditions, to distal settings
where cooler conditions prevail (Fig. 7.17). Carbonate species vary moving towards the
porphyry source from: Fe- and Mn-rich (siderite and rhodochrosite) at shallow levels, or dis-tal
to the porphyry source; to Ca- and Mg-rich (calcite, Mg-calcite, dolomite) at depth, or proximal
to an inferred porphyry source. Intermediate between these carbonate types, are the mixed Mn-
Mg-Fe-Ca carbonate species (ankerite and kutnahorite).

118
 Fig. 7.16

Zonations in vein mineralogy and styles of mineralization in base metal carbonate gold
systems. Fig. 7.17
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Limited oxygen and carbon isotope analyses on carbonates at Kelian (van Leeuwen et al.,
1990), and Porgera (Richards and Kerrich, 1993) indicate that: the Ca-rich carbonates have
strong magmatic fluid signatures, whereas the Mn-Fe carbonates are more likely to develop
from a surficial fluid origin, and the kutnahorite and dolomite demonstrate a mixing of the two
fluid sources. This zonation in carbonate chemistry is therefore interpreted to represent t he de-
scent and heating of cool low pH condensate fluids, and simultaneous mixing of these fluids
with upwelling hot magmatic dominated fluids. A similar zonation in carbonate species and
modes of formation has been documented for active porphyry-related hydrothermal systems in
the Philippines (Leach et al., 1986).

Bulk low grade gold mineralization is usually encountered in the mixed carbonate, kutna-
horite/ankerite to rhodochrosite zones, where progressive mixing between descending bicar-
bonate and upflowing mineralized fluids has taken place, commonly within dilational struc-
tural environments. High grade mineralization is encountered in restricted feeder structures
where sudden quenching of upwelling fluids has occurred close to the inferred porphyry
source.

Veining tends to be dominated by sulphides ± quartz at depth, and becomes more carbonate-
rich at the expense of these phases at progressively shallower levels. At shallow levels and/or
in outflow zones quartz locally dominates over carbonate (e.g., Karangahake, New Zealand).

Pyrite is the dominant sulphide throughout most systems. However in some deposits, pyrrhotite
becomes more abundant at depth, and is in places intergrown with magnetite (e.g., Kelian, In-
donesia). Marcasite occurs as shallow levels in many systems and pyrite locally forms collo-
form bands (melnicovite), and in rare instances is amorphous.

Base metal sulphides (commonly Zn > Pb) dominate over copper phases in most systems.
However, copper contents may increase proximal to inferred porphyry sources. Sphalerite typ-
ically contains chalcopyrite blebs and stringers, and varies from colourless to yellow (Fe-poor)
in cool distal environments, to dark red-brown to opaque (Fe-rich, marmatite) at depth. The
increase in Fe-content of sphalerite has been interpreted to be indicative of an increase in the
magmatic component to the mineralizing fluid (Simmons et al., 1988), in sulphur fugacity
(Weissberg et al., 1979), and/or temperature (Barton and Skinner, 1979).

h) Fluid flow model

Hot mineralized fluids evolve from cooling shallow level porphyry intrusives and rise along
permeable zones provided by regional structures, diatreme margins or other lithological co n-
tacts such as feeder dykes to domes, or basement plutons (Fig. 7.2). At depth, these fluids mix
with circulating meteoric waters and form gold mineralization within quartz-pyrite/arsenopyrite
vein systems, in which copper phases dominate over lead-zinc sulphides.

Gases which evolve from these upwelling fluids form gas condensate zones at surficial levels,
which are dominated by bicarbonate waters, with a minor acid sulphate component. Fluid draw-
down during the cooling of porphyry intrusives promotes the downward migration of
bicarbonate fluids prior to the exsolution of most of the metals from the parent melt. Cycling of
the hydrothermal system promotes the decent of these cool, oxygenated, moderately low pH
bicarbonate fluids deep into the hydrothermal system. Pulses of hot rising mineralized fluids
mix with these bicarbonate fluids and deposit gold mineralization within carbonate-base metal

119
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

sulphide vein/breccia systems, at various crustal levels, depending on available permeability.

In carbonate-base metal gold systems, vectors provided by alteration zonation, paragenetic se-
quence, and structure, may be used to chart the flow of both upwelling magmatic dominated
mineralized fluids, and descending bicarbonate fluids in order to target:
* high grade gold zones resulting from fluid quenching within feeder structures,
* mixing within the progressive cooling environments producing low grade bulk gold
mineralization and in which higher gold grades may develop in settings of repeated
mineral deposition,
* possible gold-copper mineralization associated with the porphyry source.

Carbonate-base metal gold mineralization passes upward to epithermal quartz gold-silver hy-
drothermal systems. In near surface outflow zones, repeated boiling and cooling of mineralized
fluids results in the formation of commonly colloform banded quartz-adularia veining. Gold
mineralization preferentially occurs in thin sulphide-rich bands or breccia zones where hot
magmatic fluids have been quenched by cool, oxygenated waters (e.g., Tolukuma, PNG, Cor-bett
et al. 1994c; Cracow, eastern Australia, section 8.vii.c.2).

i) Discussion

While carbonate-base metal gold mineralization displays characteristics of both epithermal and
porphyry deposits, systems of this type should be distinguished and treated differently during
exploration and evaluation. Both bulk mineable low grade fracture/breccia ores and higher grade
lode-style ore types are recognised. However, the former are more appropriate to modern mining
methods. The nature of fracture-controlled mineralization will be governed by the local
structural environment and should therefore be considered in the planning of the orientation of
any drilling programme, and the evaluation of that data. Dilational ore zones form at differing
orientations to the controlling structures, which may be only weakly mineralized. Drilling d i-
rections should take these angular relationships into consideration. Features such as diatreme
breccias will have a pronounced influence in any fluid flow models. Of interest is that it is pos-
sible to map out the anatomy of these hydrothermal systems using vectors described above to
define fluid flow models which aid in targeting zones of best gold mineralization and possibly
porphyry sources.

j) Examples

Some examples of carbonate-base metal systems; many of which display transitions with
quartz-sulphide, or quartz-illite- chlorite systems, are set out below.

i) Kelian, Kalimantan, Indonesia

The Kelian Mine (5.7 M oz Au), occurs within a linear zone of gold occurrences in Kaliman-
tan Indonesia (van Leeuwen et al., 1990). The position of this zone at the southern portion of an
arcuate belt of structures mapped by Pieters and Supritana (1990), and described by Mitch-ell
and Carlile (1994), is indicative of a terrain boundary. This discussion of Kelian is taken from
van Leeuwen et al. (1990).

Carbonate-base metal mineralization occurs within a sequence of Tertiary rhyolitic tuffs and
epiclastics, and overlying carbonaceous sediments, which are intruded by andesite and rhyolite
stocks as well as pre-mineral fluid (muddy) breccias, (Fig. 7.18). The muddy breccias have
been interpreted by Sillitoe (1994) as maar volcano/diatreme complexes. Hydrothermal fluids
have been focused by enhanced permeability in the shattered margins of andesite porphyry

120
Fig. 7.18

Fig. 7.20
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

intrusives, along NS (West Prampus) and NE trending structures.

Two main episodes of hydrothermal activity have been recognised at Kelian (Fig. 7.19). The
tuff/epiclastic sequence and shattered contacts of andesite plugs exhibit an early phase of quartz-
adularia-sericite + calcite veining, accompanied by intense phyllic (quartz-sericite + adularia)
alteration. Chlorite-carbonate + epidote alteration is encountered in the less permeable cores of
the porphyritic intrusive. Adularia is recognised at depth in the earlier alteration phas-es, whereas
sericite dominates at shallow levels, and persists in the later phases of this stage of activity. This
distribution of adularia-sericite is interpreted to indicate a progressive increase in gas content
with time, and condensation of gases upon boiling at shallow levels. Fluid inclu-sion analyses on
quartz and carbonate are indicative of locally boiling, mesothermal (280-350°C), relatively dilute
(< 4 equivalent percent NaCl) conditions, during this stage of hydro-thermal activity (Fig. 7.19).

Early fractures and breccias have been reactivated during the second stage of mineral deposi-
tion, characterised by the deposition of carbonates + quartz and associated base metal and gold
mineralization. Localised boiling (indicated by bladed carbonate), and associated brecciation of
earlier veining, took place in the vicinity of the blind andesitic intrusives and tension gash
fractures in the pyroclastic rocks (Fig. 7.19). The zonation from: carbonate, to carbonate +
quartz, and quartz (locally colloform banded) + carbonate, is indicative of cooling and degassing
of the fluid as it migrated westward and towards shallower levels.

Gold occurs as: inclusions in base metal and iron sulphides, intergrown with mixed element
(Mn, Fe, Ca, Mg) carbonates, or infilling fractures and cavities in earlier quartz veining. Gold
fineness ranges from 640-950, with an average of 750, typical of carbonate-base metal gold
deposits (Fig. 4.8). Fluid inclusion analyses on sphalerites and mixed element carbonates (Fig.
7.16), indicate that this later phase of activity occurred under a similar mesothermal (270-
330°C), but more saline (5 to >10 wt percent NaCl) environment, than the earlier quartz.

Carbonate species exhibit a characteristic zonation with depth, best displayed at the north end of
the deposit (Fig. 7.20). Iron (siderite) and manganese (rhodochrosite) carbonates are encoun-
tered at shallow levels, whereas magnesium-calcium (dolomite) carbonate persists at depth.
Multi-element carbonates (kutnahorite, Mg-Mn-Ca-Fe) are encountered at intermediate depths,
where mixing of hot, upwelling Ca-Mg- rich and cool, descending Fe-Mn- rich fluids occurred.
This zone of mixing typically delineates the regions of economic gold mineraliza-tion. Sphalerite
is Fe-poor at shallow levels and in the south, and progressively increases in Fe-content to
marmatitic sphalerite at depth and to the northeast.

Hotter conditions at the northern portion of Kelian are apparent from the progression from Fe-
Mn carbonates at shallow levels, to Ca-Mg carbonates and pyrrhotite at depth (Fig. 7.20). The
local intergrowth of pyrrhotite with magnetite, and change to more Fe-rich sphalerite at depth
to the north, are indicative of reducing conditions proximal to a porphyry source. Broad car-
bonate zonations to the south indicate progressive mixing resulting in low grade mineraliza-
tion, whereas telescoped narrow carbonate zones to the north are also indicative of rapid the
quenching of the hot upwelling fluids, and host local high grade mineralization. Bent and de-
formed bladed Mn-rich carbonates in these high grade zones are interpreted to illustrate rapid
quenching of boiling two phase fluids. Thus, alteration styles provide vectors of fluid flow from
an inferred porphyry source and assist in the identification of local higher grade zones.

121
Fig. 7.19
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

ii) Porgera, Papua New Guinea

The Porgera gold mine (contained gold >14 M oz) essentially represents two mineralized sys-
tems: the open pittable carbonate-base metal gold deposit and the epithermal quartz gold-silver
system extracted mainly from an underground mining operation (Corbett et al., 1995). From
initial panning of gold downstream by the first Government patrols into the district in 1938, and
subsequent alluvial miners, exploration of the bulk low grade potential at Waruwari proceeded
during the 1970's with boosts from the increase in gold price in 1980, and the dis-covery of the
Zone 7 high grade in 1983 (Henry, 1988). Production from the underground be-gan in 1990 and
the open pit in 1992. This discussion is taken from Corbett et al., (1995), Fleming et al. (1986),
Handley and Henry (1991), Richards (1990), and Richards and Kerrich (1993).

The Porgera Intrusive Complex (PIC) has been emplaced into locally calcareous shales of the
Chim Formation shelf sediments (Davies, 1983) which form part of the uplifted melange of the
New Guinea Orogen (Rogerson et al., 1987). Emplacement of the PIC at 6 Ma (Richards and
McDougall, 1990), was localised by the intersection of the NNE trending arc normal Porgera
Transfer Structure (PTS), with WNW trending structures which parallel the accretion-ary prism
(Figs. 3.1, 7.21, Corbett, 1994). One such WNW structure evident on the Wabag 1:250,000
geological map (Davies, 1983) and landsat imagery, is mapped by Porgera Joint Venture
geologists as a prominent shear to the west of Porgera (Fig 7.21, Corbett et al., 1995). Transfer
structures separate segments of the subducting plate (Fig. 3.2), and are interpreted by Hill (1990)
to locally facilitate a dextral rotation of the accretionary prism. Accretionary struc-tures display a
similar rotation across the PTS from normal to the transfer structure, and WNW on the western
side of the PTS, to NNW to the east (Fig. 7.21). A set of ENE trending structures formed normal
to the rotated eastern portion of the accretionary prism are termed ac-cretionary joints.

Gold mineralization is intimately related to the Porgera Intrusive Complex (Richards and Ker-
rich, 1993), which comprises stocks and dykes of porphyritic hornblende and augite-hornblende
diorite, locally containing olivine, and later more calc-alkaline sills and dykes of andesite and
feldspar porphyry (Fig 7.22). The outcropping intrusives are inferred to represent apophyses
capping a deeply buried magma source which radiate from a central feeder (Corbett et al. 1995).
The Waruwari intrusives do not have a magnetic root and so may have slid-off from an original
position prior to mineralization. Later cross cutting feldspar porphyry bodies are indicative of a
differentiating intrusive system from which the gold-bearing magmatic flu-ids also evolved. The
upper portion of the PIC appears to be tilted to expose the southern rim as a series of scarp
slopes while the north dipping dip-slopes are locally capped by baked sediment.

The structural elements of Porgera (Figs. 7.22, 7.23; Corbett et al., 1995), include:

* NNE trending fractures evident as landsat and air photo lineaments represent the
continuation of the deep crustal Porgera Transfer Structure through the cover of folded
and thrusted sediments. Some fractures have been mapped as shears in the mine area
and much of the early carbonate-base metal gold mineralization appears to be hosted
within NNE fractures which also appear to localise the emplacement of the later stage
feldspar porphyry stocks.

* ENE trending accretionary joint structures, which include the Roamane and parallel
faults, are inferred to have undergone dextral rotation by the regional dextral rotation
on the Porgera Transfer Structure (Corbett, 1994). Local extension associated with

122
Fig. 7.21

Fig. 7.22
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization* Corbett G J & Leach T M. 8/96 Edn.

dextral and normal fault movement on the Roamane Fault provide important ore-hosting
dilational environments for the quartz-roscoelite mineralization discussed in Sec-tion
(7.iv.d.l).

A feldspar porphyry emplaced into the intersection of a transfer structure and the Roamane
Fault, displays an inverted cone shape from which it intrudes Waruwari Hill, as well as along
the Roamane Fault and into the hanging wall splits (Figs. 7.22, 7.23). A feldspar porphyry at
Wangima may have been localised by similar intersection and migrated into the ENE fracture
mapped north of the Roamane Fault.

The two styles of alteration, veining and mineralization at Porgera are interpreted to be depos-
ited from fluids which were sourced from the same melt as late stage feldspar porphyry intru-
sions. The Stage I carbonate-base metal type of alteration veining and gold mineralization,
which is exploited in the open pit, is outlined below. The roscoelite-rich epithermal quartz gold-
silver system is described in a later Section (7.iv.d.l).

Emplacement of the PIC into incompetent Chim Formation carbonaceous to calcareous sedi-
ments resulted in the formation of a zoned brittle contact alteration which extends up to 50 -
100m from the intrusions (Figs. 7.23, 7.24). Subsequent fracturing of the brittle altered sedi-
ments provided permeability for later hydrothermal fluids. Stage I activity occurred at deep ep-
ithermal to mesothermal levels and is characterised by early quartz-sericite alteration and
veining, followed by massive sulphide mineralization and late carbonate veining (Fig. 7.25).
Fluid inclusion data (Richards and Kerrich, 1993) illustrates cooling from early quartz (aver-age
Th approx. 318°C) to later sphalerite (average Th approx. 273°C). These veins equate to the type
A and B ores of Fleming et al. (1986). The sequence of sulphide mineralization is pyrite
—> sphalerite —> galena —> chalcopyrite/tennantite, which continued into the carbonate
phase of deposition. Gold occurs as minute (typically >20-40 micron) inclusions in sulphides,
and locally as free gold in carbonate. Gold fineness (average 670, Fig. 4.8) is low for car-
bonate-base metal systems. The gold fineness decreases during progressively later phases,
which indicate that cooling occurred during Stage I veining, as supported by the above fluid
inclusion data (Richards and Kerrich, 1993). Submicroscopic gold is inferred to be associated
with localised early pyrite-arsenopyrite mineralization and equates with type C ore of Fleming
et al. (1986).

Changes in carbonate and sphalerite composition provide vectors for fluids during Stage I
veining and mineralization. There is change in the Fe-content of sphalerites, as indicated in
colour changes of dark red, red-yellow to yellow, from north to south (Fig. 7.26). This zona-tion
is interpreted to indicate higher temperatures and a greater magmatic-component (see sec-tion
7.iii.g) of Stage I fluids to the north and east of Rambari-Waruwari. This is supported by the
occurrence of pyrrhotite in the Jez Lode, northern Rambari. Copper phases are rare, how-ever
significant chalcopyrite ± magnetite mineralization at depth and to the north also provide
vectors to a magmatic source for Stage I fluids.

Carbonates are zoned from Mn- and Fe-rich (rhodochrosite ± siderite, with hypogene hema-tite)
at shallow levels and deep in major structures, through transitional Ca-Mn-Fe-Mg-rich phases
(early ankerite, and late dolomite), to Ca ± Mg-species (calcite and dolomite) at depth and to
the north (Fig. 7.26). This zonation in carbonates is interpreted to reflect the de-scent of cool,
oxygenated, possibly bicarbonate fluids down major structures during Stage I activity. Carbon
and oxygen isotope data on Stage I carbonates (Richards and Kerrich, 1993) support the
progressive mixing of magmatic and meteoric waters to produce the observed zo-nations in
carbonate species. Late Stage I calcite and dolomite are in close isotopic equilibri-um with host
calcareous sediments (Richards and Kerrich, 1993), and are

123
Fig. 7.23

Fig. 7.24
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M. 8/96 Edn.

interpreted to have been deposited from surficial bicarbonate waters which have descended, i
on waning of Stage I activity, into available open structures.

It is therefore interpreted that Stage I mineralized magmatic fluids have migrated south and
west, from the inferred feeder stock (Fig. 7.23), via the NE-trending transfer structures. These
fluids are thought to have been cooled and diluted by meteoric waters to produce low grade
gold mineralization associated with the carbonate-base metal veining; and quenched in cross-
cutting structures by cool, oxygenated bicarbonate waters and resulting in localised higher
grade. The occurrence of abundant chalcopyrite at depth beneath the East Zone at Roamane,
implies that there has also been localised migration of mineralized fluids, from the magmatic
source, south along a major NNW-trending structure (Figs. 7.25, 7.26).

iii) Morobe goldfield. Papua New Guinea

The Morobe Goldfield produced some 3.7 M oz of gold from alluvial and hard rock mining
between 1926 and 1977 (Lowenstein, 1982), and still contains substantial gold reserves at:
Kerimenge (1.8 M oz Au, Hutton et al., 1990), Hidden Valley (2.4 M oz Au, Pascoe, 1991), and
Hamata, (1.3 M oz Au, Wells and Young, 1991). Hard rock gold mineralization in the Morobe
Goldfield ranges from quartz-sulphide systems at mesothermal levels at Hamata and
Kerimenge, to carbonate-base metal style systems at mesothermal to shallow epithermal levels
at Upper Ridges, Edie Creek, Hidden Valley, and higher elevations at Kerimenge (Fig. 7.27).
The quartz-sulphide system at Hamata has been discussed previously (Section 7.ii.d).

Structural setting

The Morobe goldfield is hosted by the Bulolo Graben (Fig. 7.28), an intra-arc rift formed by
rotation on reactivated structures described by Dekker et al., (1990) as Mesozoic basement
transfer structures (Corbett, 1994). Pliocene volcanoplutonism in the environment of exten-
sional tectonism and crustal thinning facilitated emplacement in the Bulolo Graben of Edie
Porphyry flow dome complexes, and Bulolo Ignimbrite as extrusive equivalents (Fig. 7.28).
Otabanda Formation lacustrine sediments obscure the volcanic and basement rocks in the
northern portion of the graben.

Much of the gold mineralization within Bulolo Graben is controlled by graben-bounding and
intra-graben structures. At Wau, the Golden Ridges and Upper Ridges, gold mineralization
occurs in association with a maar/diatreme (Sillitoe et al., 1984), in the hanging wall of the
Escarpment Fault. The Hamata and Hidden Valley deposits similarly occur in hanging wall
settings associated with the Upper Watut graben-bounding structure. Brittle deformation
caused by the intrusion of diatreme breccias through competent basement rocks has provided
fracture vein settings for gold mineralization at Wau (Sillitoe et al., 1984), Kerimenge (Akiro,
1986) and Edie Creek (Corbett, 1994). Kerimenge is localised by the NS trending Kerimenge
Fault, an inferred splay from the transfer structures (Corbett, 1994).

Edie Creek

Much of the Edie Creek gold mineralization occurs in a 3 km long NW trending corridor of
sigmoidal, en echelon lodes (Lowenstein, 1982), adjacent to the Nauti diatreme (Fig. 7.28;
Corbett, 1994). Exposures within Nauti Creek provide a 600 m vertical section through the 5
km long diatreme, which varies from: cobble breccias containing rounded, fresh porphyry
boulders in a illite—pyrite altered matrix at the more deeply eroded central portion, to a tuff
ring breccia at the southeastern diatreme margin. The Nauti diatreme tuff ring breccias are
similar to the Namie breccias at Wau. Basement rocks exposed at an inlier in the tuff ring

124
Paragenetic sequence for Stage I event at Porgera

Fig. 7.25

Fig. 7.26
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbet! G J & Leach T M. 8/96 Edn.

breccias in Webiak Creek contain fracture-hosted gold mineralization. Some of the cobble
breccias in the western portion of the diatreme could be transitional to conglomerate units.

Patterns of vertical and horizontal carbonate zonation are apparent in the vein/lode mineraliza-
tion formed adjacent to the diatreme. Within the belt of lodes, Lowenstein (1982) records a var-
iation from the NW to SE, moving away from the diatreme margin, of: quartz-pyrite-arsenopyrite
with only very minor carbonate at the Enterprise Mine, through increasing car-bonate, to a
carbonate dominate mineralization at the Day Dawn South Mine. Gold and silver show strong
correlations with manganese contents (after manganocarbonate) with some of the highest gold
grades contained within the crustiform banded Edie Lodes. Lowenstein (1982) recognised a
paragenetic sequence of:

quartz-pyrite —> base metal sulphides —> carbonate + Ag-sulphosalts/sulphides, in banded

Mn-carbonate veins from the Karuka Mine. This sequence of deposition is characteristic of
shallow level carbonate-base metal gold systems. Gold is commonly associated with the base
metal sulphides and carbonate. The multiple deformation and mineralization evidenced by the
banding is consistent with the sigmoidal shaped lodes having developed as tension gash fea-
tures. The dilational ore hosting environment is provided by sinistral movement on NS struc-
tures inferred from the regional structural setting (Fig. 7.28; Corbett, 1994).

Further to the SE, and distal to the Nauti diatreme margin, at the Midas workings, rich gold
grades have been recorded in association with late crystalline quartz within weathered manga-
nese oxide ore. This is consistent with a lateral zonation as silica predominates towards the pe-
riphery of the system see also Karangahake, New Zealand, Section 7.iii.j).

Wau

Veining and mineralization in the Upper Ridges pit, Wau, is hosted in polyphasal quartz -
carbonate-base metal veins which crosscut brittle diatreme breccias, termed Namie Breccias
(Sillitoe et al., 1984; Fig. 7.29). These rocks have undergone post-mineral hydrothermal brec-
ciation associated with dacite dyke intrusion to form the Davidson Breccia. Two main stages of
veining and mineralization are recognised at Upper Ridges (Denwer et al., 1995; Leach, un-publ.
reports). These are:

Stage I: Quartz-sericite-pyrite veining is overgrown by massive Fe-rich (marmatitic) sphaler-ite

± galena. Fluid inclusion analyses indicate that the quartz was deposited over a wide tem-
perature range (210-390°C) and under dilute (<1 wt percent NaCl) to hypersaline (>25 wt
percent NaCl) conditions (Syka, 1985; Denwer et al., 1995). These veins have a strong bis-
muth-tellurium geochemical signature, and gold occurs as inclusions in pyrite (average fine-ness
of 613). The presence of coarse sericite wall rock alteration and high fluid inclusion tem-
peratures indicate that these veins have been emplaced into the breccias at depths of >500-1000
m below the paleosurface. These conditions are comparable to those which formed quartz-
sulphide auriferous veining at Ribroaster (Syka, 1985) at higher elevations at the inter-section of
a cross structure with the Escarpment Fault (Fig. 7.28).

Stage II: Polyphasal Carbonate-base metal-quartz veining grades from early banded Mn-
carbonate through massive carbonate to late quartz ± carbonates. Minor pyrite, Fe -poor
sphalerite, galena and chalcopyrite are intergrown with both carbonate and quartz. Traces of
late stage tin-phases (canfieldite and stannite) overgrow the base metal sulphides. Low fine-
ness gold (average 468) occurs as inclusions, with base metal sulphides, in pyrite. Stage II
quartz and carbonate were deposited under much cooler conditions (198-220°C) than the early

125
Fig. 7.27

Fig. 7.28
 Exploration Workshop "Southwest Pacific rim gold-coppsf systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

quartz-sulphide veining (Denwer et al., 1995; Syka, 1985).

At lower elevations from Upper Ridges, the Wau maar-diatreme complex occurs as a circular
feature which is rimmed by endogenous dacite domes and infilled by epiclastic and pyroclastic
material (Sillitoe et al., 1984). The presence of very low temperature smectite-kaolinite-
cristobalite alteration, current hydrothermal solfataras on the diatreme margins, and reported
historical hydrothermal eruptions from Koranga Crater, all suggest that the Wau diatreme-maar
complex is very recent. Silica sinters and travertine deposits aligned along the Wondumi Gr a-
ben Structure (Fig. 7.28) also appear to be related to current geothermal activity in the Wau
region.

It is therefore interpreted that hydrothermal activity at Wau occurred over a protracted period and
that alteration and mineralization were focused along the Escarpment Fault, and possibly related
to emplacement of an Edie Porphyry at depth. Quartz-sulphide gold mineralization oc-curred
early and deep in the system, whereas carbonate-base metal style of gold mineralisation took
place later, and possibly at shallower levels. The Escarpment Fault possibly facilitated
emplacement of a recent high level Edie Porphyry, and associated formation of the Wau maar-
diatreme complex and endogenous dacite domes. Ejecta on the margins of the Upper Ridges pit
are inferred to have been derived from eruption of the Wau diatreme and cover material de-rived
from an older source. Rocks similar to the Namie breccias occur along the Escarpment fault
(Denwer and Leach, unpubl data) and west on the margin of the Nauti diatreme (Fig.
7.28). Large allochthonous blocks of quartz-carbonate-sulphide veined Namie Breccia (Gold-en
Peaks and Golden Ridges) have slid into the Wau maar, possibly facilitated by recent
movements on the Escarpment Fault.

Kerimenge

The Kerimenge prospect is localised at the intersection of the NS trending Kerimenge Fault
with a diatreme breccia (Fig. 7.28, Corbett, unpubl. map, 1985; Akiro, 1986). Fracturing at this
structural intersection provides an important focus for fluids derived from an inferred porphyry
source at depth. Ore-hosting NW trending and SW dipping fractures are inferred to have formed
as hanging wall splits, by sinistral rotation and normal faulting on the Kerimenge fault. Denwer
et al., (1995) note that the best gold grades occur at the intersection of the frac-ture/veins and
the controlling Kerimenge Fault (Fig. 7.30).

Two main stages of hydrothermal activity have been recognised at Kerimenge (Fig. 7.30, Syka
and Bloom, 1990; Denwer et al., 1995):
Stage I: Low Grade, refractory, polyphasal quartz-sulphide comprises:
* Early quartz-sericite-pyrite veining and locally intense silicification which has
overprinted zoned porphyry-related biotite/potassic and propylitic alteration.
* Fracturing and brecciation of the silicified zones and deposition of
arsenopyrite-pyrite-marcasite-quartz. This is the main gold mineralizing event with a
resource of 51 Mt at 1.0 g/t gold, in which the gold is refractory.
* late manganocarbonate-illitic clay-arsenopyrite/pyrite + base metal sulphides.

The quartz-sulphide veining took place under low temperature (145-240°C) and dilute (<3.3 wt
percent NaCl) conditions, in which the environment of deposition became progressively cooler.
Pre- or early-Stage I quartz contain daughter phases indicative of hypersaline porphyry-related
conditions. The Stage I veining and alteration are zoned from: quartz-sulphide at depth and to
the south, through quartz-manganocarbonate-illite/sericite at interme-diate levels, to
manganocarbonate lodes at high elevation and to the north (Fig. 7.30).

126
Erosion of Upper Ridges - Ribroaster Systems offset by faulting in Escarpment Fault;
Subsequent emplacement of Wau-Karanga Crater Maar-Diatreme Complexes.

Fig. 7.29

Fig. 7.30
Fig. 7.31

Fig. 7.32
 Exploration Workshop "Southwest Pacific rim gold-copper systems: Structure, Alteration, and Mineralization" Corbett G J & Leach T M, 8/96 Edn.

Stage II: High grade, non-refractory mineralization co-exists with manganocarbonate veining,
and comprises quartz-carbonate-sulphide deposition in dark bands alternating with Mn-
carbonate, or in breccia zones sealing clasts of Mn-carbonate veining. High fineness (average
837) gold is non-refractory and is associated with hessite, tennantite, chalcopyrite mineraliza-
tion. Fluid inclusion data (Denwer et al., 1995) indicates that mineralization has taken place in
response to the mixing of cool (160-170°C) bicarbonate waters related to manganocarbonate
deposition, and hot, but dilute (<2.1 wt percent NaCl) upwelling fluids. Gold mineralization at
Kerimenge has changed from a refractory quartz-sulphide style system early and at deep levels
in the south prospect area, to a non-refractory carbonate-quartz-base metal sulphide event at
shallow levels in the northern prospect area. It is interpreted that Stage II metal-bearing fluids
are related to a separate intrusive event than Stage I mineralization.

Hidden Valley

The Hidden Valley deposit (2.4 M oz Au, Pascoe, 1991) occurs as stockwork veining devel-oped
in the hanging wall of the Hidden Valley Fault (Nelson et al., 1990), inferred to represent a
continuation of the Upper Watut Graben Fault extending from the Hamata deposit (Fig. 7.28).
Thus, the hanging wall setting is similar to Wau and Kerimenge. Nelson and co-workers de-
scribe the following paragenetic sequence of overprinting veins and fractures:

1. Quartz-pyrite, with chlorite, epidote

2. Quartz-hematite-chlorite-pyrite; rare gold with arsenopyrite
3. Carbonate with
(a) quartz-adularia
(b) base metal sulphides-gold-adularia
(c) kutnahorite-gold-tetrahedrite

Early quartz-pyrite-arsenopyrite with rare gold veining is comparable to the early quartz-
sulphide veining at Kerimenge. The sequence of quartz-adularia —> base metal sulphide-gold
—> Mn/Mg-carbonate - gold is similar to Stage II/III base metal/carbonate events recognised in
other carbonate-base metal systems (Fig. 7.15).

iv) Woodlark Island, Papua New Guinea

Gold was discovered at Woodlark Island, 300 km east to the mainland of Papua New Guinea, in
1895 (Stanley, 1912). Much of the gold production of 100,000 oz from lodes and 83,000 oz
from alluvials occurred prior to World War II (McGee, 1978). Although recent coralline lime-
stone deposits obscure much of the geology of Woodlark Island, an analysis of the aeromag-
netic data suggests: a central horst block is transected by NW trending cross structures (Fig.
7.31), may be tilted to the north, and most areas of gold mineralization occur within erosional
inliers in the cover (Fig. 7.31; Corbett et al., 1994a). Two main mining centres are at Ku-
lumadau and Busai.

Busai

Between 1902 and 1916 the Murua United Mine (known locally as the Busai Pit) produced about
3500 ounces of gold at a grade of 4.3 g/t Au and with an average fineness of 771-846 (McGee,
1978). The Busai Pit is localised at the intersection of a NS structure with a series of NW trending
cross structures, which transect a horst block (Figs. 7.31, 7.32). Mineralization occurs in the hang-
ing wall to the arcuate shape NW-dipping the Blue Lode shear where demagnetisation of primary
magnetite in the host Okiduse Volcanics is indicative of clay alteration (Fig. 7.31).

127
Exploration Workshop "Southwest Pacific rim gold-copper systems : Structure. Alteration and Mineralization" Corbett GJ & Leach TM. 3/96 Edn.

The bulk of the gold mineralization is hosted in carbonate vein/breccias within shallowly dipping
ten-sion gash features, constrained between the steeply dipping NW structures (Figs. 7.34, 7.35).
Car-bonate vein/breccias also exploit the structures utilised by the milled matrix fluidised
breccias. High grade lodes mined by the early miners and intersected in drilling, are interpreted
to represent the fis-sure veins which act as feeder structures for the shallow dipping tension gash
veins (Corbett et al., 1994a).
A paragenetic sequence of overprinting structure, alteration and mineralization events has
been de-fined (Figs. 7.35; Corbett et al., 1994b) as:
1. Propylitic alteration of volcanic pile and development of hematitic fracture and breccia
fillings, interpreted to be of deuteric origin.
2. Intrusive milled matrix fluidised breccias fine upwards from coarse grained and angular at
depth to "flinty" chalcedonic silica/pyrite fault fill at higher levels in the system. These
exploit pre-existing structures and provide ground preparation for later mineralization.
3. Polyphasal quartz-pyrite grades from early very fine chalcedonic quartz-pyrite to later
coarse grained quartz-pyrite-illitic clay-carbonate, and is equivalent to Stage I quartz
described for other systems. Local jasperoid reflects shallow oxygenated environments.
Fluid inclusion data indicates late drusy quartz was deposited from a dilute (<2 wt percent
NaCl) two phase (boiling) system at a temperature of around 280°C.
4. Gold mineralization is hosted within carbonate vein/breccias. Fluid inclusion data (Fig. 7.16)
and the vertical zonation in carbonate type from Fe/Mn at shallow levels, to mixed Mn/Mg/Fe/Ca
at intermediate levels, and to Ca/Mg-carbonate at depth, are interpreted to be indicative of mixing
hot (>250°C), relatively saline (>6 wt percent NaCl) fluids, with cool (<150°C) dilute (<2 wt
percent NaCl) bicarbonate condensate fluids. This gradual mixing formed the bulk low grade gold
mineralization as carbonate tensional gash veins. Bonanza gold grades developed as carbonate-
quartz breccias within the lodes or feeder structures, and are interpreted to result from the
quenching of upwelling mineralized fluids by ground waters (Fig. 7.33). Rare base metals are
overgrown by early quartz and form pre- to syn-carbonate deposition. Gold occurs as electrum
mainly intergrown with carbonate, but also as inclusions in sulphides, and has an average fineness
of 830, which is unusually high for carbonate-base metal gold mineralization (Fig. 4.8).
The zonation in the carbonate type from Mn/Mg carbonate at Busai to Mn carbonate at the
Federa-tion workings, some 500 m along strike to the north, is consistent with a fluid source to
the south (Fig. 7.31). The Muniai area lies in the centre of a circular feature inferred to represent
a caldera collapse feature, defined from the aeromagnetic data (Fig. 7.31). Here, lower
temperature propylitic alteration (actinolite-chlorite- albite-epidote-carbonate) is coincident with
an aeromagnetic high, pos-sibly indicative of magnetite bearing higher temperature propylitic or
potassic alteration at depth. Phyllic alteration with a low temperature clay overprint are evident
at the adjacent Bomagai pro-spect (Fig. 7.31). Thus, Busai and the other gold occurrences (e.g.,
Woodlark King, Little McKenzie, etc; Fig. 7.31), which form a roughly circular distribution,
may have been derived from a central porphyry source.
Kulumadau
Between 1901 and 1950 mining at Kulumadau produced 77,000 ounces of gold at an average grade
of 15.9 g/t Au and fineness of 776-859 (McGee, 1978). The Kulumadau mining centre is lo-calised
on a major horst-bounding structure, possibly by the intersection of a cross cutting struc-ture (Fig. 7.
31). Although similar styles of mineralization and overprinting relationships are recog-nised at
Kulumadau and Busai (Corbett et al., 1994a), the intense post mineral shearing

128