Proceeding MGEI 2016

PROOCEDINGS OF UNCONVENTIONAL EXPLORATION TARGETS &
LATEST TECHNIQUES AND NEW TOOLS IN MINERAL AND COAL

EXPLORATION
th
MGEI 8 ANNUAL CONVENTION 2016
5-6 October 2016, Bandung, West Java, Indonesia

Masyarakat Geologi Ekonomi Indonesia (MGEI) Ikatan Ahli Geologi Indonesia (IAGI)
Editor in Chief : Arif Zardi Dahlius

Alison M Purba
Herryadi Wachyudin
Paper Reviewer : Emmy Suparka

Bakti Harahap
Noel White
David Cooke
Nurcahyo Indro Basuki
Lucas D Setiyadji
STJ Budi Santoso
Cover Design by Dede Noor Iman Taufik

ii
PREFACE
It is with great pleasure that I welcome you on behalf of the Board of Masyarakat Geologi Ekonomi
Indonesian Society of Economic Geologists (MGEI) to the eight Annual MGEI Convention in
Bandung, which themed Unconventional Exploration Targets & Latest Techniques and New Tools
in Mineral and Coal Exploration, and present to you the proceedings of this event to share
knowledge, experience, update the activities in exploration, development, technology including
introduction to new concepts and ideas of metallic mineral and coal including UCG deposits.
This event is one of a series of annual events of MGEI. Series of this event consists of three sessions;
pre-convention workshop, seminar and field trip to Krakatoa.
We believe that we have invited the best experts in the field where they will share their rich and
extensive knowledge and experience with us; all that we believe will deliver and bring enormous
benefits to the world of economic geology in Indonesia and the surrounding region
We warmly thank all authors and reviewers, and the editorial board for their invaluable
contributions. We also thank our industry sponsors who made publication of the proceedings
possible. We furthermore acknowledge and thank management of the companies who allowed
sharing of knowledge of their projects. We are particularly grateful to the Organizing Committee
whose hard work and dedication have played a major part in making this convention a success.
We look forward to meeting you again at the 9th MGEI conference in 2017!
Arif Zardi Dahlius

MGEI Chairman
iii
Table of Contents
PREFACE
Table of Contents Paper Title
Table of Contents Author
The decline in Indonesias Exploration
Performance-what can we do about it
Exploring in lithocaps and green rocks new
methods to aid discovery of porphyry and
epithermal deposits
Successful Techniques in Exploring the Lithocap
Environment of the Sunda Magmatic Arc,
Indonesia
Orogenic Gold Deposits - From Hypozonal to
Epizonal Systems
Advances in understanding various ore deposits
in ultramafic rocks in Indonesia
Characterisation of Coal Deposits for
Underground Coal Gasification
Successful application of spectral analysis in
lithocap exploration and ore deposit discoveries
New Determination of alteration zonations using
SWIR in Kyisintaung high sulfidation copper
deposit, Monywa district, Central - Western
Myanmar
Ore Pathfinders and Fertility Assessments in
Lithocaps: a Case Study of the Pascua-LamaVeladero High-Sulfidation Epithermal Au-Ag
district, Argentina and Chile
The structure evolution of Zijinshan high
sulfidation epithermal deposit, Fujian Province,
China
Drones (UAVs) in mining and exploration. An
application example: Pit Mapping and
Geological Modelling.
The Power of Geology Model Supported by 3D
Geophysical Inversion and Database System as
A Tool to Generate Exploration Target in
Mined Exploration Area: Case Study of COW
A Geology Model at PT Freeport Indonesia
Fuzzy Logic and Weight of Evidence Approach
for Mineral Discovery Tool
Geobotanical Remote Sensing for Mineral
Exploration in Thick Vegetation Areas
An update of key characteristics of Awak Mas
mesothermal gold deposit, Sulawesi Island,
Indonesia
Discovery, Geology and Origin of the Lakuwahi
Volcanogenic Au-Ag-Pb-Zn Deposit, Romang
Island, eastern Indonesia.
iii
iv
v
1 Richard Schodde
David R. Cooke, Noel White, Lejun Zhang,
Michael Baker, Nathan Fox, Jennifer
Thompson, J. Bruce Gemmell, Jamie
5
Wilkinson, Pete Hollings and Huayong
Chen
Adi Maryono, Rachel Harrison, Iryanto
7
Rompo, Eddy Priowasono and Malcom
Norris
15 Franz Michael Meyer
19 Ade Kadarusman
23 Dr. Len Walker
29 Rachel L. Harrison
35 Yi Sun1 and Jing Chen

Lejun Zhang, Noel White, David R. Cooke,
Huayong Chen, Francisco Testa, Aldo
39
Vsquez, Simon Griffiths, Jennifer
Thompson, and Mike Baker
43
Jing Chen, David R. Cooke, Jos Pique,

Lejun Zhang
45
Krisztin Szentpteri, Tatzky Reza

Setiawan, and Arief Ismanto
51 Nur Wiwoho and Wahyu Sunyoto
61 Harman Setyadi, Budi Santoso, STJ

N. H. Hede, K. Koike, K. Kashiwaya, S.
Sakurai, R. Yamada, and D. A. Singer
Ernowo, Franz Michael Meyer, Arifudin
75
Idrus, Heru Widyanarko and Nita Lely
Endrasari
67
79 Joseph Ogierman
iv
The metamorphic rock-hosted gold

mineralization at Bombana (Southeast
Sulawesi) and Buru Island (Maluku): Their key
features and significances for gold exploration
in Eastern Indonesia
Lesson Learnt from Nickel Sulfide Exploration in
Latao Block, Southeast Sulawesi: A Geological
and Geometallurgical Perspective
Beneficiation and Mineralogical Study of Bauxite
Deposits in Ketapang, West Kalimantanfor
Optimum Bayer Alumina Refinery Process
Magmatic Fe-Ni-Cu Sulphides Occurence in
Sebuku Island
Investigation on Scandium and REE potential in
Sulawesi, Indonesia: a geochemical constraint
Uranium Exploration in Sulawesi
Eksplorasi Logam Tanah Jarang Di Jalur Timah
Indnesia
HVSR passive seismic surveying for mineral,
coal or shallow quarry materials exploration:
three mineral exploration case studies.
Arifudin Idrus1*, Sukmandaru

Prihatmoko2, Ernowo Harjanto3, Franz
85
Michael Meyer4, Irzal Nur5, Wahyu
Widodo3 & Lia Novelia Agung3
95
Budhi Kumarawarman, Gde Handojo

Tutuko, and Moh. Erwin Syam Noor
103
Robby I. Rafianto, Henry A. Cahyono,

Abimanyu Yudhaswita, Alan Matano,
Yusni Marta
113 Yoseph C. A. Swamidharma

Adi Maulana, Kenzo Sanematsu, Masayuki
Sakakibara
119 I G. Sukadana, and H. Syaeful
117
123 Setiawan Raharjo

133 David Stannard, Jayson Meyers
PROCEEDINGS
Unconventional Exploration Target & Latest Technique and New Tools in Mineral and Coal Exploration
Bandung, West Java, 5 - 6 October 2016
The decline in Indonesias Exploration Performance-what can we do about it

Richard Schodde
Managing Director of MinEx Consulting Pty Ltd and Adjunct Professor at Centre of Exploration Targeting University of
Western Australia.
ABSTRACT
Historically, Indonesia has been very successful at exploration - and many of the discoveries have led to major new mines
for the country. While exploration expenditures doubled over the last decade, the number of discoveries fell by more than
half. Consequently, the cost per discovery has increased four-fold from $52m in the 1986-2005 to ~$200m in the last
decade (all in constant 2015 US Dollars). Of major concern is that the current lack of discoveries will cause a decline on
Indonesia mine production in the longer term.
There are several reasons for the decline in discovery performance many of which also apply to other countries around
the World. However, Indonesia also has its own set of special problems such as the UIP Moratorium, issues over Forest
Permits and the ban on exporting unprocessed minerals.
The solutions to the above are fairly self -evident. However, their implementation is challenging. Even so, the long term
economic benefits to the country from fixing this are immense.
Coal, >10 Mt Coking Coal, >10 Mt Fe, >10 Mt Al2O3 or
its equivalent.
INTRODUCTION
Indonesia has a long history of mining - dating back
over one thousand years (Van Leeuwen, 1994). However,
large scale exploration and mining really only took off in
the 1970s. In detail, half of all of Indonesias known
significant deposits (totaling 260 to date) have been found
since 1983.
Of significance and concern is the observation that,
in spite of large expenditures on exploration, the rate of
discovery has slowed downed in the last decade (see
Figure 1).
This paper reviews the causes behind the decline in
discovery performance and suggests possible ways to
improve it.
Fig. 2 Size, type and location of significant mineral deposits in

Indonesia Note: The different bubble sizes refers to the size of
the deposit (Moderate, Major and Giant)
Of these 10,334 deposits, 260 (or ~2.5%) are located

in Indonesia (see Figure 2). To put this into perspective
Indonesia accounts for just 1.4% of the Worlds land
mass.
The 260 significant deposits are made up of 127 bulk
mineral deposits (mainly coal) and 133 non-bulk deposits.
This includes 69 gold deposits and 43 base metal deposits
and 21 deposits of other minerals.
With regard to the 133 non-bulk mineral deposits, the
author has information on the discovery dates for 125 of
these deposits.
The number of discoveries were then compared
against the level of exploration expenditures as reported
SNL and MEG in their annual surveys of the industry
(SNL & MEG, various years). This survey started in 1997,
However it should be noted that it only covers non-bulk
exploration activities. Consequently, the author limited
his analysis of the Indonesias discovery performance to
consider only non-bulk minerals.
Fig. 1: Exploration expenditures and number of discoveries in

Indonesia
Note: Analysis excludes expenditures and discoveries of bulk
mineral deposits (such as coal, bauxite and iron ore)
DATA AND METHOD

The following analysis is based on a database
compiled by MinEx Consulting, which has information on
over 55,700 mineral deposits around the World, including
10,334 deposits that are >= moderate in size. This
defined as deposits containing >0.1 Moz Au, >5 kt U3O8,
>10 kt Ni, >0.25 Mt Zn+Pb, >0.1 Mt Cu, >20 Mt Thermal
RESULT AND DISCUSSION

One hundred non-bulk deposits have been found in
Indonesia since 1950, of these 44 have been converted
into mines; and the average delay between discovery and
development was 9.7 years. The 44% conversion rate is
1
PROCEEDINGS
due to high fixed costs (for staff, offices and access fees),
much of the cut-back has been directed to less field work.
However, if you dont drill, you wont discover.
Some of these challenges are cyclical (and so should
improve over time), whereas others (such as the depth of
cover issue) are structural.
Indonesia also faces additional challenges of its own
making. These include;
1) UIP Moratorium which creates uncertainty over the
security of title as existing tenements reach the end of
their lives
2) Restrictive Forestry classifications which may
contain prospective areas for exploration
3) Delays in issuing Forestry Permits which slows
down the companys ability to carry out field
activities
4) Requirement for minerals to be processed in-country.
For many metals, and companies with small projects,
this adversely impacts on the project economics
5) Divestment
requirements
for
foreign-owned
operations. The current arrangement appears to result
in mines being sold for less than fair value, thereby
impacting on the incentive to explore in the first-place
As a result, Indonesias share of the Worlds
exploration expenditures has halved from 2.1% in 2012 to
1.1% in 2015. Over the same time period the number of
foreign companies actively exploring in Indonesia
declined from 51 to 35.
It should be noted that the foreign companies play a
key role in discovery. Over the last 30 years they
accounted for 65% of the countrys non-bulk mineral
discoveries.
According to the latest survey of mining companies
active in Indonesia (Fraser Institute, 2016), the four main
issues inhibiting investment in the country are;
1) The legal system
2) Regulatory duplication
3) Administration of regulations, and
4) Disputed land claims
Historically the exploration industry has generated
considerable wealth for the country. It is estimated that
over the period 1986-2005, the industry created $5.00
worth of wealth for every Dollar spent on exploration.
However, over the last decade (2006-2015), this has
dropped to just $0.61 per Dollar spent. In other words,
exploration is currently not generating wealth for the
investors. This is driven by the lack exploration success
particularly for the (high-value) Tier-1 and -2 deposits.
For the long-term prosperity of Indonesia, it is
imperative that the Government find ways to stimulate
exploration and discovery in the country.
The solutions to the above are fairly self -evident.
However, their implementation can be challenging.
In the first instance, efforts should be made to
encourage foreign companies come back to Indonesia as
they have historically found most of the countrys
deposits. Simultaneously, efforts should be made to
comparable to the World average rate of 45% over the

same period; and Indonesias delay period is significantly
shorter than the World average of 12.4 years (Schodde,
2014).
With regard exploration expenditures, over the two
decades between 1986 to 2005 a total of US$2.81 billion
(in constant 2015 US Dollars) was spent on exploring for
non-bulk minerals in Indonesia. This led to the discovery
of 54 significant deposits, at an average cost of $52
million per discovery. However, over the last decade,
from 2006 to 2015, industry spent $2.33 billion on
exploration but only found 10 deposits. Even after
factoring the potential for additional (as yet) unreported
discoveries, the average unit discovery cost over the last
decade is in excess of $200 million per deposit. In other
words, even though annual exploration expenditures
increased by a factor of 2, the annual discovery rate
halved resulting in a four-fold increase in discovery
costs. As indicated in Figure 3, the average cost is
currently around $267 million per discovery, up from a
low of just $24m back in 2005.
Fig. 3: Average cost per mineral discovery in Indonesia and

Western World. ---Note: Analysis is based on a 4-year rolling
average, and is based on non-bulk mineral deposits only.
Figure 3 shows that, a similar, albeit less dramatic,

increase in discovery costs also occurred elsewhere in the
Western World. There are many reasons behind the
decline in discovery performance in both Indonesia and
other countries; these include:
1) Higher input costs associated with the mining boom.
For example, in Australia, between 2000 and 2012 the
cost (as measured in constant US Dollars) of hiring a
geologist and drilling a hole increased by 150% and
125% respectively. Office administration costs also
rose by 170% of the same time period. Similar trends
were observed in other countries (Schodde, 2013).
2) Shift towards exploring under deeper cover, with a
commensurate increased difficulty in identifying
suitable targets for drilling
3) Shift away from greenfield to brownfield exploration;
with the latter less likely to deliver large new
discoveries
4) Decline in the general quality of the exploration
targets. During the boom-times good money was
often wasted exploring marginal projects, and
In the current downturn, many companies have
significantly reduced their exploration budgets. However,
2
PROCEEDINGS
encourage local companies to take a more active role in

exploration.
Over the longer term, efforts should be made to
improve the skills base of local geologists and provide
companies with the necessary pre-competitive data (such
as better maps, and more detailed geophysical and
geochemical surveys) to find the next generation of
discoveries.
To do this the Government needs to re-engage with

the mining industry. In particular it needs to encourage
local companies to become much more active in
greenfield exploration. At the same time the Government
needs to regain the trust and support of the foreign
exploration companies- as they have historically been
source of most of the deposits found in Indonesia.
While the implement of such policies may be
challenging, the long term benefits to the countrys
prosperity are immense.
CONCLUSIONS
The international mining industry is facing several
challenges on many fronts and central to this is the need
to improve its discovery performance. In addition,
Indonesia has its own special challenges and
opportunities.
Given the fact that only around half of all discoveries
are converted into mines, and that there is (on average) a
decade delay between discovery and development,
Indonesia is facing a real risk that mineral production (and
associated revenues) declining in the future.
To redress this imbalance, the Government needs to
put in-place policies that will stimulate exploration and
boost the number of significant discoveries.
REFERENCES
Fraser Institute, 2016, Survey of Mining Companies,
published by the Fraser Institute, Vancouver.
Schodde, R.C., 2013, The impact of changes in gold price on
exploration activities and strategies presented at the
NewGenGold Conference, Perth.
Schodde, R. C., 2014, Key issues affecting the time delay
between discovery and development, presented at PDAC
Convention March 2014.
SNL and MEG, various years, Corporate Exploration
Strategies, annual survey as published by SNL and the
Metals Economics Group, Halifax.
Van Leeuwen, T. M, 1994, Journal of Geochemical Exploration,
50, 13-90
PROCEEDINGS
PROCEEDINGS
Exploring in lithocaps and green rocks new methods to aid discovery of porphyry
and epithermal deposits
David R. Cooke1,2, Noel White1,3, Lejun Zhang1,2, Michael Baker1,2, Nathan Fox1, Jennifer Thompson1,2, J. Bruce
Gemmell1,2, Jamie Wilkinson4, Pete Hollings5 and Huayong Chen1,6
1
CODES, the Australian Research Councils Centre for Excellence in Ore Deposits, University of Tasmania, Private Bag
79. Hobart, Tasmania, 7001, Australia
2
Transforming the Mining Value Chain, an ARC Industrial Transformation Research Hub, University of Tasmania,
Private Bag 79. Hobart, Tasmania, 7001, Australia
3
Ore Deposit and Exploration Centre (ODEC), School of Resources and Environmental Engineering, Hefei University of
Technology, Hefei 230009, Anhui, China
4
Department of Earth Science and Engineering, Imperial College London, Exhibition Road, London SW7 2AZ, United
Kingdom
5
Geology Department, Lakehead University, 955 Oliver Road, Thunder Bay, Ontario, Canada P7B 5E1
6
Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, 511 Kehua Street, Tianhe, PO Box 1131,
Guangzhou, China, 510640.
ABSTRACT
Porphyry-related copper, gold and molybdenum deposits and related deposits continue to be major exploration targets for
many exploration companies. For cost-effective exploration, field campaigns need to maximize the information that can be
garnered from early exploration stages so that deposits are discovered quicker and at less cost by reducing the amount of
drilling required for discovery. This is particularly relevant in regions of poor outcrop (e.g., jungle, desert, young cover),
where an added disincentive to continue after initial results can be management fatigue. Geochemical exploration tools are
particularly challenged in tropical weathering environments, where geochemical dispersion during weathering can obscure
or destroy the geochemical anomalies associated with porphyry and epithermal mineralisation.
CODES, and in recent times the TMVC Industrial Transformation Research Hub, and their international research partners
have conducted a series of AMIRA International research projects aimed at developing new geochemical and geological
techniques for the exploration of lithocaps and green rocks associated with porphyry and epithermal deposits. A series of
four AMIRA projects have been conducted over the past twelve years: AMIRA P765 (Transitions and Zoning in Porphyry Epithermal Districts: Indicators, Discriminators and Vectors: 2004 2006), P765A (Geochemical and Geological Halos in
Green Rocks and Lithocaps Developing the Explorers toolbox: 2008 2010), P1060 (Enhanced Geochemical Targeting
in Magmatic-Hydrothermal Systems: 2011 2014) and P1153 (Applying the explorers toolbox to discover porphyry and
epithermal Cu, Au and Mo deposits: 2015 2018). These projects have developed and validated new geochemical
techniques that can help to predict the likely direction and distance to mineralised centres (vectors), either in the deeply
eroded green rock environment, where propylitic alteration predominates, or in the intensely clay-altered lithocap
environment, where porphyry or high sulfidation-style epithermal mineralisation may be hidden in a sea of intense clay and
silicic alteration.
Primarily based on advances in laser ablation-inductively coupled plasma microanalysis, the AMIRA International
research programme has demonstrated that subtle hypogene geochemical dispersion halos can be detected several
kilometres beyond the limits of geochemical anomalies that are detected by conventional rock-chip sampling techniques,
thereby extending the detectable geochemical footprint of porphyry mineralised centres. The team also developed
geochemical tools that discriminate between environments (porphyry, epithermal, metamorphic), and made significant
progress towards developing and testing fertility indicators (i.e., discriminating large, small and barren systems). In the
lithocap environment, a combination of short wavelength infrared analyses with high quality whole rock geochemical data
and mineral chemistry analyses of alunite, quartz and/or pyrite can help to identify domains of high temperature alteration
that are prospective for porphyry and/or high sulfidation epithermal mineralisation. In the green rock environment, chlorite
and epidote are now well-established as tools that aid in vectoring and fertility assessments. Magnetite, quartz, tourmaline
and calcite also show promise as aids to exploration in more deeply eroded porphyry-epithermal districts. In this
presentation, we will demonstrate how the combination of epidote and chlorite chemistry with grass-roots geology can be
used to help successfully detect the location of a giant, high-grade porphyry deposit.
PROCEEDINGS
PROCEEDINGS
Successful Techniques in Exploring the Lithocap Environment of the Sunda

Magmatic Arc, Indonesia
Adi Maryono1, Rachel Harrison2, Iryanto Rompo3, Eddy Priowasono4 and Malcom Norris5
1
J Resources, Equity Tower, 48th Floor, SCBD, Jl. Jend Sudirman, Kav 52-53, Lot 9, Jakarta, 12190 Indonesia. Email:
adi.maryono@jresources.com
2
Independent Consultant Geologist, Borobudur
3PT Buena Persada
4
PT Newmont Nusa Tengggara
5
Avalon Minerals Limited
ABSTRACT
This paper attempts to document successful exploration techniques that have been employed during recent discoveries of
major Au-Ag (+5 Moz Au) and Cu-Au (+10 Moz Au and +5 Mt Cu) deposits concealed beneath barren or mineralized
lithocap bodies along the Sunda magmatic arc. There are at least four discoveries of potentially large porphyry Cu-Au
deposits in the last 10 years including Elang in Sumbawa in 2005 and Tumpangpitu in East Java in 2009. Tumpangpitu is
the most recent discovery, containing 28 Moz Au and 8.6 Mt Cu. Recent major discoveries with economic potential have
been reported from Brambang in Lombok in 2010 and Huu in Sumbawa in 2015.
The integration of state of the art exploration techniques (geology, geochemistry and geophysics), and excellent teamwork
have been thought to be key ingredients to exploration success. BLEG (bulk leach extractable gold) drainage geochemical
surveys have proven to be more effective than other stream samples (-#80 silt and pan concentrate/PC samples) to locate
sizeable altered areas at regional-district scales and early exploration stages. A combination of lithocap-porphyry
mapping-logging techniques and application of spectral data (ASD) has been very powerful in providing vectors to ore, and
for delineation of both epithermal and porphyry targets in the Elang, Tumpangpitu and Brambang districts. Both mapping
techniques involve observation of aspects of intrusions, alteration mineralogy, sulfide mineralogy, hypogene oxides, vein
types and density, supergene Fe-Cu oxides, rock textures, magnetic susceptibility measurements, and structures.
Delineation of the lateral surface extent of lithocaps and understanding alteration zoning is crucial, as they can be used as
a proxy suggesting the size of the underlying mineralized porphyry intrusions. Key spectral parameters towards causative
intrusions include increasing crystallinity and changes in composition (wavelength) of white micas, position of the
wavelength feature between 1480 and 1490 nm for alunites (Na/(Na+K) ratio/high Ca alunite), high Fe/(Fe+Mg) ratios of
chlorites and distribution of iron oxides.
The combination of soil geochemistry (low level Mo anomalies) and ground-airborne magnetic data (small discrete subtle
magnetic anomalies) has provided direct tools to delineate mineralized porphyry zones and to define drill targets under
large lithocap bodies at Tumpangpitu. The latest 3D modeling of the airborne magnetic data indicates potassic-altered
porphyry targets at depth at Elang, Tumpangpitu, Brambang and Huu. Application of 3D inversion pole-dipole resistivity
modeling and airborne time domain electromagnetic (HoisTEM) and magnetic surveys has been very successful in mapping
lithocap bodies as well as surface mineralized quartz ledges at Martabe and concealed porphyry targets at Elang.
delineation of drill targets resulting in world class
discoveries.
INTRODUCTION
This paper emphasizes successful exploration
techniques that have been employed during recent
discoveries of major Au-Ag and Cu-Au deposits
concealed beneath barren or mineralized lithocap bodies
along the Sunda magmatic arc. Major porphyry Cu-Au
deposits that crop out at Batu Hijau, Elang and Selodong
were discovered in the early 1990s, but porphyry Cu-Au
targets concealed under large lithocaps along the same
magmatic arc have only been identified during the past 10
years.
Recent discoveries of world-class Au-Ag (+5 Moz
Au) and Au-Cu deposits (+10 Moz Au and +5 Mt Cu)
along the Sunda magmatic arc and elsewhere have been
attributed to new understanding of lithocap formation and
state of the art exploration techniques. Previous explorers
have failed to identify potentially large resources, due to
barren to weakly mineralized surface geochemical and
weak geological signatures of large lithocaps. Integration
of geological, geochemical and geophysical techniques, as
well as good teamwork have played a key role in the
RECENT MAJOR DISCOVERIES OF WORLD-CLASS

AU AND CU-AU DEPOSITS
Major discoveries of Au-Ag and Au-Cu deposits
during the last 15 years have placed the Eastern Sunda
magmatic Au-Ag-Cu belt as one of the most fertile belts
in the world with a total metal endowment of 92.44 Moz
of Au, 279.17 Moz of Ag and 61.92 Blb (28.1 million
tonnes) of Cu (Maryono et al., 2012). It contains three
world-class porphyry Cu-Au deposits at Batu Hijau, Elang
and Tumpangpitu (+10 Moz Au and +5 Mt Cu) together
with significant high- sulfidation epithermal Au-Ag
deposits (+0.5 Moz Au).
Along the Sunda magmatic arc, major discoveries
include Martabe in Sumatra in 1999 (Sutopo et al., 2003;
Harlan et al., 2005), Elang in Sumbawa in 2005 (Maryono
et al., 2005) and Tumpangpitu in East Java in 2009
(Norris et al., 2011; Harrison and Maryono, 2012).
Tumpangpitu is the most recent discovery, containing 1.9
billion tonnes @ 0.45% Cu and 0.45 g/t Au containing 28
7
PROCEEDINGS
Moz Au and 8.6 Mt Cu (Intrepid Mines Ltd., 2012).

Recent major discoveries with economic potential have
been reported from Brambang in Lombok in 2010 and
Huu in Sumbawa in 2015. There are at least four
discoveries of potentially large porphyry Cu-Au deposits
in the last 10 years in the Eastern Sunda magmatic arc.
Tumpangpitu had previously been explored by
several mining companies in the 1990s, but the large
resource potential of a porphyry Cu-Au deposit was only
recognized in 2010 (Fig. 2; Norris, 2011). A world-class
resource of porphyry Cu-Au mineralization at Elang was
identified and delineated in 2005, or 15 years after the
discovery of porphyry Cu-Au mineralization (Maryono et
al., 2005). It took more than 20 years of exploration to
identify the large resource potential of porphyry Cu-Au
deposits concealed under thick lithocaps at Brambang,
SW Lombok and Huu in Sumbawa.
Hijau, the porphyry deposit is exposed at surface, with

only deep remnants of the lithocap remaining. At Elang,
Sumbawa, and Selodong Lombok, partial preservation of
a lithocap is apparent. At Tumpangpitu, East Java,
Brambang, Lombok and Huu, Sumbawa, entire lithocaps
are preserved
with high-sulfidation epithermal
mineralization intact at Tumpangpitu. Blind porphyry
targets are concealed under barren overlying large lithocap
bodies at Brambang and Huu at depths of ~150 m and
~200 m below current surface respectively.
A full spectrum of the vertical extents of lithocaps
along with underlying porphyry deposits for the Eastern
Sunda magmatic arc has been compiled from several
epithermal and porphyry deposits; this can used as a
conceptual exploration model for the region (Fig. 3).
High-sulfidation epithermal Au-Ag mineralized systems
are developed within lithocaps associated with
quartzalunite-enargite ledges at Elang, Selodong and
Tumpangpitu. Quartz-enargite veins with advanced
argillic selvages penetrate to depths exceeding 1 km in the
porphyry Cu-Au mineralized zone at Tumpangpitu,
forming a deeply telescoped system (Harrison and
Maryono, 2012).
The porphyry Cu-Au deposits with overlying large
lithocaps along the eastern Sunda arc are spatially
associated with small Neogene nested dioritic to tonalitic
intrusive complexes with low-K calc-alkaline to weakly
alkaline dioritic to tonalitic compositions. Intrusion ages
range from 2.7 Ma at Elang, 3.7 Ma at Batu Hijau to 7.5
Ma at Selodong (Garwin, 2002; Maryono et al., in press).
Mineralizing intrusive bodies consist of multiple phases;
early, intermediate and late tonalite intrusions with latest
intrusive activity marked by post-mineralization diatreme
breccia bodies that are developed along the margins or
adjacent to the porphyry systems and partially or totally
disrupt the mineralized bodies. The tonalite intrusions are
elongate, with pencil-like geometries 200 m to 500 m in
diameter and with >2 km vertical extent. The porphyries
are hosted by or intrude along the margins of coarsegrained, equigranular dioritic stocks and batholiths. The
depth of emplacement of the porphyry intrusions ranges
from 1 to 2 km to 5 km below the paleo surface (Maryono
et al., 2012).
Supergene copper mineralization is only developed
beneath goethitic leached caps at Batu Hijau and Elang. A
weak chalcocite blanket averaging 40 m thick and 0.5 to
0.7 % Cu (500 m by 750 m in plan view) with overlying
goethite-hematite leached cap at the surface has been
intercepted in drill holes at Elang (Maryono et al., 2012).
LITHOCAP GEOMETRY AND EXPLORATION MODELS

Understanding the geometry and formation processes
of lithocaps is crucial, as a number of world-class
porphyry Cu-Au deposits and epithermal Au-Ag deposits
along the Sunda arc are associated with large lithocaps
(+20 km2). Lithocaps were recognized in the 1990s as
large horizontal to subvertical blankets of residual quartz
and advanced argillic alteration of hypogene origin above
intrusions (Sillitoe, 1995; White, 1991; Hedenquist et al.,
1998). The areal extent of lithocaps can be >20km2 and
they are temporally and genetically related to intrusions.
Improved understanding of lithocaps in the 2000s has
triggered further exploration in the Sunda magmatic arc.
The vertical extent of lithocaps was discussed by Sillitoe
and Hendenquist (2003), to demonstrate relationships
between high- and intermediate-sulfidation epithermal
mineralization and underlying porphyry deposits. Detailed
studies of exploration tools in the lithocap at the
Mankayan Au-Cu district, Luzon Philippines have been
documented by Chang et al. (2011), to demonstrate the
genetic link between the Far southeast porphyry Cu-Au
deposit and the Lepanto high-sulfidation epithermal AuCu deposit, hosted in a large lithocap body covering an
area of 7km by 4 km, and the immediately underlying
Victoria and Theresa intermediate-sulfidation epithermal
Au-Ag deposits.
The lateral extent of lithocaps associated with major
porphyry Cu-Au and high-sulfidation epithermal Au-Ag
deposits along the Eastern Sunda magmatic arc are
commonly large, varying from 1 x 0.3 km at Batu Hijau
(mapped as feldspar-destructive alteration; Clode et al.,
1999), 10 x 3 km for Tumpangpitu (Harrison and
Maryono, 2012), 6 km x 4 km at Elang and Brambang
(Maryono et al., 2012) to 8 km x 6 km at Huu
(Pratiwinda, 2015 pers. communication). The dimensions
of these lithocaps are in part due to differing erosion
levels and exposures of the porphyry deposits. At Batu
SUCCESSFUL EXPLORATION TECHNIQUES LEADING TO

MAJOR DISCOVERIES
At regional and district scales and early exploration

stages, BLEG (bulk leach extractable gold) drainage
geochemical surveys have proven to be an effective
8
PROCEEDINGS
2005) and Tumpangpitu (Harrison and Maryono, 2012;

Harrison, 2014). Spectral data were further integrated
with logging data to construct geological models and to
separate signatures of the porphyry hydrothermal system
from the overprinting high sulfidation epithermal system
(alteration zoning and ore characterization) at
Tumpangpitu. Key spectral parameters that vector towards
causative intrusions include increasing crystallinity and
changes in composition (wavelength) of white micas,
position of the wavelength feature between 1480 and 1490
nm for alunites (Na/(Na+K) ratio/high Ca alunite), high
Fe/(Fe+Mg) ratios of chlorites and distribution of iron
oxides.
An airborne time domain electromagnetic survey
(HoisTEM) survey has been flown to map marginal
argillic, proximal advanced-argillic and central siliceous
lithocap footprints at Martabe and Elang. This EM survey
along with 3D inversion modeling of ground pole-dipole
IP surveys has proven to be superior to and provide deeper
penetration than the preceding gradient array surveys both
at Martabe and Elang (Hoschke, 2011). In addition, the
latest airborne magnetic surveys indicated the lithocap
bodies within a broad magnetic low containing small
discrete and subtle magnetic anomalies inside; these
indicate potassic-altered porphyry targets. 3D magnetic
modeling has been used to predict the geometry of
concealed porphyry targets at depth at Elang,
Tumpangpitu, Brambang and Huu.
The mineralized quartz-alunite ledges at Martabe are
clearly defined by the pole-dipole high resistivity values
(+20 ohm-m to +3000 ohm-m) that show a strong
correlation to lithology, alteration and gold grades (0.5 to
+5 g/t Au) as shown from deposit modelling at Purnama
and plan depth slices (Sutopo et al., 2003; Hoschke,
2011). Strong correlation of the mineralized quartz-alunite
ledges and high resistivity bodies (+20 ohm-m to 1,000
ohm-m) can also be clearly seen from depth slices of the
3D HoisTEM modeling (Hoschke, 2011). The lithocap
zones at Ladam-Jelatang and Gerbang in the Elang district
correspond to highly resistive bodies clearly defined
within the conductive zones (+15 ms to +50 ms) which are
consistent with 3D inversion of ground pole-dipole
IP/resistivity data. Latest airborne magnetic surveys have
also confirmed a concealed southern porphyry target at
Elang under the lithocap as initially observed during the
mapping using the lithocap-porphyry mapping technique.
technique to locate sizeable altered areas. One sample can

cover a catchment area of more than 20 km2, which is
more effective than other stream samples (-#80 silt and
pan concentrate/PC samples). However, a lack of initial
integration of anomalies with geological and geophysical
datasets failed to identify concealed porphyry targets
underneath lithocaps at Tumpangpitu, Brambang and
Huu (Fig. 3).
The magnitude of BLEG geochemical signatures
reflect erosional levels from deeply eroded to fully
preserved systems. Strong BLEG geochemical anomalies
correspond to outcropping porphyry-epithermal systems
(e.g., Batu Hijau, Elang and Martabe), moderate
anomalies to partly exposed systems (e.g., Tujuh Bukit
district and Selodong) and weak anomalies are associated
with concealed porphyry systems (e.g., Brambang and
Huu). BLEG gold anomalies decrease downstream over 8
km, as seen at Batu Hijau from 196 ppb Au near the
deposit, to 7 ppb at the southern coast (Sejorong and
Tongoloka streams; Maula and Levet, 1996). Similar
BLEG geochemical dilution is shown at Martabe from
100 ppb near the Purnama deposit, to 14 ppb Au at Aek
Pahu River, a distance of 5 km (Sutopo et al., 2003). In
contrast, only weak BLEG gold signatures (2 to 10 ppb
Au) occur around the mineralized centers, with no
anomalous values in -80# silt and PC samples associated
with barren lithocaps at the surface or ~150 m above
concealed porphyry targets at Brambang, Lombok.
Lithocap mapping techniques have been successfully
employed to delineate surface footprints of lithocap bodies
and to target high-sulfidation epithermal systems at Elang
(Maryono et al., 2005). This mapping technique can be
used to define vectors to ores at shallow levels, from
marginal argillic zones to proximal advanced-argillic
zones and central mineralized residual quartz-alunite
ledges. Porphyry mapping techniques (modification of the
Anaconda mapping method) provides further tools to
vector to the central mineralized porphyry intrusions at
Elang and Brambang. Both techniques involve
observation of aspects of intrusions, alteration mineralogy,
sulfide mineralogy, hypogene oxides, vein types and
density, supergene Fe-Cu oxides, rock textures, magnetic
susceptibility measurements, and structures. Observation
of key breccia aspects (geometry, matrix, fragments,
textures, structures, juvenile magmatic clasts, mineralized
clasts, bedded and massive facies) is also important to
determine vectors and to indicate concealed mineralized
targets at depth. This is particularly valuable if postmineralization diatreme breccias are present, as at Batu
Hijau, Elang, Tumpangpitu and Brambang.
The application of vis-NIR and SWIR data from
the use of analytical spectral devices (ASDs, PIMA) has
been instrumental in providing spectral vectors toward
mineralized centers when integrated with surface
geological mapping and logging at Elang (Maryono et al.,
SUMMARY
The key ingredients to exploration success along the
eastern Sunda magmatic arc have been attributed to the
integration of geology, geochemistry and geophysics, and
excellent teamwork. These successes have employed
systematic exploration approaches and applied state of the
art exploration techniques and used experts in multiple
disciplines. Delineation of the lateral surface extent of
9
PROCEEDINGS
Porphyry and Epithermal Deposits: Example from the

Mankayan Intrusion-Centered Cu-Au District, Luzon,
Philippines, Society of Economic Geologists, Economic
Geology v.106, p.1365-1398.
Clode, C.H. Proffect,J.M. and Munajat, I., 1999, Timing
relationship of intrusion, wall-rock alteration, and
mineralization in the Batu Hijau copper-gold porphyry
deposit. Proceedings Pac-Rim Congress, 10-13 October
1997, Bali,pp. 485-498.
Garwin, S., 2002, The geologic setting of intrusion-related
hydrothermal systems near the Batu Hijau porphyry coppergold deposit, Sumbawa, Indonesia. Society of Economic
Geologists, Special Publication 9: 333-366
Harrison, R.L., and Maryono, A., 2012, Tumpangpitu porphyryhigh sulfidation epithermal deposit, Tujuh Bukit
Project, Indonesia - geology, alteration and mineralization:
Red Metals Symposium Presentation, CODES,
University of Tasmania, 51 p.
Harrison, R.L., 2014, The application of short wave infrared
(SWIR) spectral analysis in exploration for the
Candrian porphyry copper-gold-molybdenum and highsulfidation epithermal prospect, Tujuh Bukit district,
southeast Java, Indonesia: Master of Economic Geology
thesis, University of Tasmania, 126 p.
Hedenquist, J.W., Arribas, A., Jr., and Reynolds, T.J., 1998,
Evolution of an intrusion-centered hydrothermal system;
Far Southeast-Lepanto porphyry and epithermal Cu-Au
deposits, Philippines: Economic Geology, v.93, p.373-404.
Hoschke, T., 2011, Geophysical signatures of copper-gold
porphyry and epithermal gold deposits, and implication for
exploration, Codes Arc center of Excellence in Ore Deposits
University of Tasmania, 47p.
Intrepid Mines Ltd, 2012, Intrepid Mines Limited: Porphyry
Resource Model Announcement OCT 9, 2012 - 08:36 ET
http://intrepidmines.com.au/investor-relations/news-andannouncements/
Maryono, A., Lubis, H., Perdankusumah, A. and Hermawan, W.,
2005, The Elang Porphyry Copper and Gold Mineralization
Style Sumbawa, Indonesia, Indonesian Minerals and Coal
Discoveries, 19 20 September 2005, IAGI.
Maryono, A., Setidjaji, L.D., Arif, J., Harrison, R.L., and
Soeriaatmadja, E., 2012, Gold, Silver and Copper
Metallogeny of the eastern Sunda Magmatic Arc, Indonesia:
MGEI Annual Conference, Proceedings of Banda and
Eastern Sunda Arcs 2012, p.23-38.
Maula, S. And Levet, B.K., 1996, Porphyry copper-gold
signatures and the discovery of the Batu Hijau deposit,
Sumbawa, Indonesia, in Porphyry Related Copper and Gold
Deposits of the Asia Pacific Region: Cairns, Queensland,
Australian Mineral Foundation, Conference Proceedings (
August 12-13, 1996), pp. 10.1-10.3.
Newmont Mining Corporation, 2012, Profitable Growth with
Diciplined
Returns,
http://newmont.q4web.com/files/doc_presentations/Diggers
%20%20Dealers%20v5.pdf
Norris, M., 2011, The Discovery History of the Tujuh Bukit
Copper-Gold Project East Java, Indonesia, in NewGen Gold
2011 Conference Proceedings: Gold Mining Journal and
Louthean Media Pty. Ltd., Perth, Western Australia.
Sillitoe, R.H., 1995, Exploration of porphyry copper lithocaps:
Australian Institute of Mining and Metallurgy, Punblication
Series, v.9, p.527-532.
Sillitoe, R.H., and Hedenquist, J.W., 2003, Linkages between
volcanotectonic setting, ore-fluid compositions and
epithemal precious-metal deposits: Society of economic
Geologists Special Publication 10, p.315-343.
Sutopo, B., Jones, M.L. and Levet, B.K., 2003, The Martabe
gold discovery: A high sulfidation epithermal gold-silver
deposit, north Sumatra, Indonesia, in NewGen Gold 2003
lithocaps and understanding alteration zoning is crucial, as

they can be used as a proxy suggesting the size of the
underlying mineralized porphyry intrusions. For the past
10 years, discovery of major porphyry Cu-Au potential at
Brambang and Huu, and delineation of world-class
porphyry Cu-Au resources at Elang and Tumpangpitu are
the product excellent multi-disciplinary teamwork and
fine exploration work. Batu Hijau was discovered earlier
in the 1990s from outcropping porphyry exposures and
strong geological, geochemical signatures that were easier
to recognize.
A combination of lithocap-porphyry mapping
techniques and application of spectral data (ASD) has
been very powerful in providing vectors to ore, and for
delineation of both epithermal and porphyry targets in the
Elang and Brambang districts. This work followed
successful location of prospective areas by regional and
infill BLEG drainage surveys, which have been shown to
be superior to traditional silt and PC stream surveys. The
combination of soil geochemistry (low level Mo
anomalies) and ground-airborne magnetic data (small
discrete subtle magnetic anomalies) has provided direct
tools to delineate mineralized porphyry zones and to
define drill targets under large lithocap bodies at
Tumpangpitu. Application of 3D inversion pole-dipole
resistivity modeling and airborne EM and magnetic
surveys has been very successful in mapping lithocap
bodies as well as concealed porphyry targets at depth.
ACKNOWLEDGEMENTS
It has a pleasure and privilege to have worked with
excellent geoscientists and exploration teams at Batu
Hijau, Elang, Tumpangpitu and Brambang that have
developed a wealth of knowledge, skills and produced
such successful results. We would like to thank the
following individuals; Brian Levet, Chris Clode, Hasbi
Lubis, the late Bruce Harlan, Wawan Hermawan, Johan
Arif, Dudy Setyandhaka, Arif Perdana, John Proffett and
Terry Hoschke during Newmonts period; Malcolm
Norris, Bruce Rohrlach, Damien Lulofs, David Gray,
Andreas Kristianto and Chris Moore during Intrepid
times; Syafrudin Maula, Bambang Irianto, Rayes
Sembiring, Mido and Widya during exploration at
Brambang.
Our gratitude also goes to those who have made
major contributions toward discoveries as so many
geoscientists and supporting staff involved that we cannot
mention all their names. They remain an important part of
our history. It is all about people! Our thanks and respect.
Special thanks go to Noel White for his review and
valuable inputs. Well keep bothering you, Noel.
REFERENCES
Chang, Z., Hedenquist, J.W., White, N.C., Cooke, D.R., Roach,
M., Deyell, C.D., Garcia, J. Jr., Gemmell, J.B., McKnight,
S., and Cuison, A.L., 2011, Exploration Tools for Linked
10
PROCEEDINGS
Conference Proceedings: Gold Mining Journal and

Louthean Media Pty. Ltd., Perth, Western Australia.
White, N.C., 1991, High sulfidation epithermal gold deposits:
Characteristics and a model for their origin: Geological
Survey of Japan Report 277, p. 9-20.
11
PROCEEDINGS
Fig. 1. Major Au-Ag and Au-Cu deposits recently discovered along the Sunda magmatic arc, including the most recent, Tumpangpitu
(containing 30.1 Moz Au and 19 billion lbs of Cu; Intrepid Mines Ltd., 2012; Maryono et al., 2012).
Fig. 2. Drilling in 2010 (GTD-10-163) intersected a large concealed porphyry Cu-Au deposit under a large lithocap body at Tumpangpitu,
just 100m away from previous drill hole GTD-001 drilled in 1997 (Norris, 2011)
12
PROCEEDINGS
Fig. 3. A schematic section showing conceptual exploration model of overlying lithocap with high-sulfidation Au-Cu deposits over
porphyry Cu-Au deposits along the Sunda magmatic arc (modified from Maryono et al., 2012)
Fig. 4. Integration of geology, soil geochemistry (low level Mo anomalies) and geophysical magnetic data (small subtle discrete magnetic
anomalies) has been very powerful in the delineation of altered areas and locating drill hole targets at Tumpangpitu (modified from
Norris, 2011)
13
PROCEEDINGS
14
PROCEEDINGS
Orogenic Gold Deposits - From Hypozonal to Epizonal Systems

Franz Michael Meyer
Institute of Applied Mineralogy and Mineral Deposits, RWTH Aachen University, Wllnerstrasse 2, 52056 Aachen,
Germany. Present address: German-Mongolian Institute of Resource Technology, GMIT Campus, 2nd khoro Nalaikh
district, Ulaanbaatar, Mongolia.
ABSTRACT
This paper summarizes research carried out on orogenic gold deposits by staff and students of the Institute of Mineralogy
and Economic Geology, RWTH Aachen University. The review focuses on mineralization signatures of 11 mines and mining
districts essential to the depiction of this class of gold deposits. Mineralization ages range from 3027 Ma to 6-8 Ma, but the
age distribution is not continuous with a gap over 1.5 Ga between the late Archaean Pilgrims Rest district and the
Neoproterozoic Navachab deposit. Most of the deposits investigated are older than 2 Ga.
Based on P-T conditions of ore formation, the deposits are grouped into mesozonal and hypozonal systems. In relation to
stages of metamorphism, plutonic activities, and tectonic settings gold mineralization is generally syn- to late-deformation.
Mesozonal deposits are typically syn- to post-peak greenschist facies whilst hypozonal types is related to syn-to late-peak
amphibolite facies metamorphism. Two of the investigated mines display two distinct stages of gold mineralization.
Intrusion-related deposits are not directly related to regional metamorphism. Sources for metals and Au-transporting fluids
both are thought to include mainly metamorphic and minor magmatic as well in two cases of mixed metamorphic-magmatic
environments.
The salient question that will be addressed is, if there are common traits that occur systematically in deposits from different
terrains having different ages. In other words, are there definitive characteristic similarities or differences in mineralization
processes and Au-transporting fluids that were responsible for their formation?
The orogenic gold deposits portrayed here have a high areal concentration in Archaean greenstone belts and the questions
remains is, what are the implications for defining new target areas for an orogenic gold deposits along Phanerozoic
metamorphic belts?
not continuous as our data set reveals a gap over 1.5 Ga
between the late Archaean Pilgrims Rest district and the
Neoproterozoic Navachab deposit. Most of the deposits
investigated are older than 2 Ga.
INTRODUCTION
This paper reviews research carried out on orogenic
gold deposits by staff and students of the Institute of
Mineralogy and Economic Geology, RWTH Aachen
University (references [1]-[12]). Case studies include the
following mines and mining districts (in order of
decreasing mineralization ages): New Consort, Barberton
Greenstone Belt, South Africa; Cuiab, Rio das Velhas
Greenstone Belt, Brazil; Ajanahalli, Chitradurga
Greenstone Belt, India; Renco, Limpopo Belt, Zimbabwe;
Hutti/Hira, Hutti-Maski Greenstone Belt, India; Pilgrims
Rest, Transvaal Supergroup, South Africa; Lega Dembi,
Megado Belt, Ethiopia; Navachab, Damara Belt, Namibia;
Mindyak, Urals, Russia; Kochkar, Urals, Russia; Awak
Mas, Sulawesi Island, Indonesia.
Figure 1. Ages of Mineralization
DEFINITE ORE DEPOSIT CHARACTERISTICS
Pressure-temperature conditions of mineralization are

shown in Figure 2. The fields for hypozonal, mesozonal
and epizonal orogenic gold deposits are delineated. In the
figure is also included geothermal gradients as well as wet
solidi for granite, pelite, and gabbro.
The diagram reveals that our data does not include
epizonal deposits and that the temperature distinction
between mesozonal and hypozonal mineralization is in the
region of the temperature range of the main metamorphic
fluid production in mafic bulk compositions. Renco and
New Consort formed at pressures in excess of 4 kbar, but
pressure conditions at Hutti, Navachab and the second
mineralization stage at New Consort overlap with those of
the mesozonal deposits.
The review focuses on mineralization signatures

essential to the depiction of orogenic gold deposits. They
include ages of mineralization, P-T conditions of ore
formation and relation to stages of metamorphism and
plutonic activities, tectonic settings, associated host rocks,
fluid chemistry, mineral associations, and styles of
hydrothermal alteration. The salient question that will be
addressed is if there are common traits that occur
systematically in deposits from different terrains having
different ages. In other words, are there have definite
similarities characteristic or different in mineralization
processes and Au-transporting fluids that were responsible
for their formation?
Mineralization ages summarized in Figure 1 range from
3027 Ma to 6-8 Ma. However, the distribution of ages is
15
PROCEEDINGS
New
(HZ)
Consort
Cuiab
Ajjanahalli
Table 1 indicates the host rock types of the investigated

orogenic gold deposits. In general, they may be hosted by
all lithologies present in the local environment. Favorable
hosts are mafic and ultramafic volcanic rocks (New
Consort, Cuiab, Hutti), Fe-rich gabbroic sills (Pilgrims
Rest), granitoid intrusions (Renco, Kochkar), as well as
banded iron formation (Cuiaba, Ajjanahalli), and
siliciclastic metasedimentary rocks (Lega Dembi,
Pilgrims Rest, Navachab, Mindyak, Awak Mas). Gold
mineralization is predominantly confined to (often
laminated) quartz-carbonate vein networks or may be
present within Fe-rich sulfidized wallrock selvages, or
silicified and arsenopyrite-rich replacement zones. There
is no distinctive differentce in host rocks between
hypozonal and mesozonal deposit types.
Host Rock Lithology
Mafic-ultramafic volcanics
BIF; mafic volcanics
Ajjanahalli
BIF
Renco
Enderbite
Hutti (HZ)
Amphibolite; felsic schist
Pilgrims's Rest
Shale, dolomite, gabbroic sills
North Limpopo thrust zone
Hutti (HZ)
Hutti-Maske shear zone
Pilgrims's Rest
Synkinematic mafic intrusion
Siliciclastic sediments
Marble; calc-silicates; biotite
Navachab (HZ)
shist
Mindyak
Tectonic melange
Composite granitoids
Awak Mas
Flysch-type meta-sediments
shear
Navachab (HZ)
Mon Repos thrust zone
Mindyak
Main Uralian fault
Kochkar
Synkinematic granitoids
Awak Mas
Synkinematic intrusions?
Deposit
New Consort
(HZ)
Cuiab
Ajjanahalli
Renco
Hutti (HZ)
Structurally epigenetic hypozonal and mesozonal

deposits are generally hosted in deformed metamorphosed
terranes (Table 2). Most deposits are distributed along
major compressional to transtensional crustal-scale fault
zones in deformed greenstone terranes. Gold
mineralization is localized in moderately to steeply
dipping, compressional brittle-ductile shear zones and
faults as shallow-dipping extensional veins and
hydrothermal breccias. Intrusion-related deposits are
associated with granitoids and mafic intrusions.
Pilgrims's Rest
Table 2: Structural setting

Deposit
Lega-Dembi-Aflata
zone
Table 3: Timing of mineralization relative to metamorphism
Lega Dembi
Kochkar
Chitradurga boundary shear

zone
A salient question for the formation of epigenetic

orogenic gold deposits is the timing of mineralization
relative to deformation and metamorphism. Table 3
indicates that gold mineralization is generally syn- to latedeformation. Mesozonal deposits are typically related with
syn- to post-peak greenschist facies while hypozonal types
formed in the syn-to late-peak amphibolite facies
metamorphism. New Consort and Hutti display two
distinct stages of gold mineralization. Intrusion-related
deposits are not directly related to regional
metamorphism.
When considering typical hydrothermal alteration
assemblages associated with orogenic gold mineralization
a distinction is to be made between mesozonal and
hypozonal types. Mesozonal mineralization in association
with greenschist facies metamorhism is generally
accompanied by a proximal carbonate and sericite
alterations a whilst distal zone is characterized by chlorite.
In contrast, hypozonal alteration can be more complex and
varied in amphibolites facies terranes.
Table 1: Host rock lithology (HZ = hypozonal)

Deposit
New Consort
(HZ)
Cuiab
Cuiab tubular sheath fold
Renco
Lega Dembi
Figure 2: Pressure-temperature conditions of mineralization
(modified from [12])
Consort Bar shear zone
Timing relative to
Metamorphism
2
stages,
post-peak
metamorphism
syn- to late-peak metamorphism
post-peak metamorphism
2
stages,
post
metamorphism
intrusion-related
peak
Lega Dembi
syn- to late-peak metamorphism
Navachab (HZ)
syn-peak metamorphism
Mindyak
Kochkar
peak-metamorphism
Awak Mas
intrusion-related?
Common assemblages associated

with
gold
mineralization include garnet, biotite, amphibole,
diopside, K-feldspar, plagioclase, calcite, and titanite.
Host Structure
16
PROCEEDINGS
Alteration minerals seem to inherit their chemical

signature from the host rocks and the alteration
assemblages are in general thermal equilibrium with the
country rocks. There is a distinct mineralogical difference
between 1st and 2nd mineralization stages at New Consort
and Hutti, most likely related to the prevailing differences
in P-T conditions.
Mindyak
Kochkar
Awak Mas
Orogenic gold mineralization is typically associated

with low salinity (generally < 3-7 wt% NaCl equiv.), CO2H2O-rich, near neutral pH hydrothermal fluids. They are
further characterized by the presence of H2S, CH4 and/or
N. However, despite the advent of high-resolution
analytical equipment that can measure metals and isotopic
compositions in single fluid inclusions the long prevailing
controversy about the source of ore fluids responsible for
orogenic gold mineralization still continuous. Single fluid
inclusion analyses of halogens, metals, and even stable
and radiogenic isotopes do not provide unequivocal
evidence of the fluid source.
However, geological evidence, age data, and isotope as
well as geochemical studies favor a metamorphic model
for fluid and metal sources. Fluids are thought to be
produced by metamorphic dehydration of for example
chlorite (12% H2O) or amphibole (2% H2O) and
decarbonation reactions. There is no general association
between orogenic gold deposits and magmatic intrusions.
However, some deposits seem to be intrusion-related and
overlap in age with magmatic events. In this case a mixed
magmatic-hydrothermal model may be most appropriate
for explaining the source of metals and fluids. This
ambivalence related to the source is also displayed in out
data set (Table 6). Cuiab, Ajjanahalli, Renco, Lega
Dembi, Mindyak, and Kochkar are thought to have a
metamorphic fluid source. The mixed metamorphicmagmatic model seems to be more appropriate for New
Consort, Hutti, and Navachab, while Pilgrims Rest and
probably Awak Mas appear to be intrusion-related.
Table 4: Hydrothermal alteration mineralogy

Deposit
Alteration Assemblage
1st stage: garnet, diopside, hornblende, KNew
feldspar, quartz, calcite, biotite.
Consort
2nd stage: hornblende, plagioclase, K-spar,
(HZ)
biotite, titanite, quartz
chlorite,
carbonate,
sericite,
quartz,
Cuiab
zoisite/clinozoisite
Ajjanahal
chlorite, stilpnomelane, minnesotaite, sericite,
li
ankerite
Renco
garnet, biotite, K-feldspar, quartz
Hutti
1st stage: biotite, chlorite, plagioclase. 2nd
(HZ)
stage: chlorite, K-feldspar
Pilgrims's
ferroginous carbonates, quartz, chlorite,
Rest
sericite, rutile
Lega
actinolite, biotite chlorite, epidote, calcite,
Dembi
sericite
Navachab
1st stage: garnet, diopside, quartz, K-feldspar.
nd
(HZ)
2 stage: garnet, biotite
Mindyak
quartz, albite, sericite, chlorite, carbonates
mafic dykes: biotite, actinolite, albite, Kfeldspar, quartz, epidote,
Kochkar
tourmaline, sericite; granitoids: quartz,
sericite, calcite, epidote
Awak
quartz, albite, sericite, carbonate
Mas
The main gangue minerals in orogenic gold lodes are

quartz and carbonates together with variable amounts of
sericite, chlorite, amphibole, biotite, scheelite, and
tourmaline. Main ore minerals include native gold, pyrite,
pyrrhotite, and minor amounts of chalcopyrite,
arsenopyrite, tellurides, and bismuth minerals (Table 5).
Sulfides typically constitute less than 10% of the ore with
generally very low base metal sulfides. Significant vertical
mineral zonation appears to be absent. The assemblage
arsenopyrite together with loellingite is conspicuous in
hypozonal deposits where arsenopyrite is the main sulfide.
Table 6: Source of ore fluids

Deposit
New Consort
(HZ)
Cuiab
Cuiab
Ajjanahalli
Renco
Hutti (HZ)
Pilgrims's
Rest
Lega
Dembi
Navachab
(HZ)
Fluid Source
Mixed magmatic-metamorphic
Metamorphic
Ajjanahalli
Metamorphic (O-, S-isotopes)

Metamorphic (O-, S-, SrRenco
isotopes)
Hutti (HZ)
(O-, H-, S-, B-isotopes)
Magmatic (O-, H-, S-, C-, SrPilgrims's Rest
isotopes)
Lega Dembi
Metamorphic (Sr-isotopes)
Navachab (HZ)
(O-, H-, C-, S-isotopes)
Mindyak
Metamorphic
Table 5: Ore mineralogy

Deposit
New
Consort (HZ)
bismuth tellurides
pyrite, arsenopyrite, chalcopyrite, sphalerite
pyrite, pyrrhotite, arsenopyrite, galena,
tetradymite, tellurobismuthite
pyrite, chalcopyrite, galena
Ore Mineralogy
pyrrhotite, arsenopyrite, loellingite
pyrite, pyrrhotite, arsenopyrite, chalcopyrite,
sphalerite
Pyrite, pyrrhotite, marcasite, arsenopyrite,
chalcopyrite, sphalerite
pryrite, pyrrhotite, chalcopyrite, sphalerite
molybdenite, cubanite,
magnetite ilmenite rutile, native bismuth,
bismuth alloys
pyrite, pyrrhotite, arsenopyrite, chalcopyrite
pyrite, arsenopyrite, fahlores, chalcopyrite,
bismuthinite, bismuth
pyrite,
pyrrhotite,
arsenopyrite,
chalcopyrite, galena, gersdorffite, sphalerite,
tellurides, niccolite, bournonite, silvertetrahedrite, molybdenite
pyrrhotite,
chalcopyrite,
sphalerite,
arsenopyrite, bismuth, bismuthinite,
Kochkar
Metamorphic (O-, Sr-isotopes)
Awak Mas
Magmatic? (O-, S-isotopes)
I. CONCLUSIONS
The above summary of salient properties of 11
orogenic deposits allows formulating a number of
statements: There are common attributes that occur
systematically in deposits from different terranes with
different
ages.
Orogenic
gold
deposits
are
characteristically associated with deformed and
17
PROCEEDINGS
Harjanto, E., Meyer, F.M., Idrus, A. (2016). Hydrothermal

alteration and gold mineralization of the meta-sedimentary
rock hosted gold deposit Awak Mas, Sulawesi Island,
Indonesia. 35th IGC Cape Town, South Africa.
Kolb,J., Dziggel, A., Bagas, L. (2015). Hypozonal lode gold
deposits: A genetic concept based on a review of the New
Consort, Renco, Hutti, Hira Buddini, Navachab, Nevoria
and The Granites deposits. Precambrian Research, 262, 20
44.
Wilkinson, B.H., Kesler, S.E. (2010). Tectonic-Diffusion
Estimate of Orogenic Gold Resources. Economic Geology,
105,1321133.
metamorphosed terranes, particularly in spatial association

with major crustal structures. Spatial and temporal
associations with granitoids of various compositions are
observed in some deposits, which may indicate that melts
and fluids were both inherent products of thermal events
during orogenic processes. The data suggest metamorphic
to magmatic as well, become as mixed metamorphicmagmatic sources for the Au-transporting fluids. Orogenic
gold deposits have formed over more than 3 billion years
of Earth's history, episodically during the Middle Archean
to Neoproterozoic times, and continuously throughout the
Phanerozoic.
What are the implications for defining new target areas
for the class of orogenic gold deposits? A recent estimate
of orogenic gold resources [13] based on age-frequency
and spatial density calculation assumes that the areal
concentration of orogenic gold deposits along Phanerozoic
metamorphic belts is about ten times that in Proterozoic
terranes and about three times higher than in Archaean
belts. The authors [13] suggest that exploration for the
next generation of orogenic gold deposits should be
focused on Phanerozoic terranes, such as those occurring
in Indonesia.
REFERENCES
Otto, A., Dziggel, A., Kisters, A.F.M., Meyer, F.M. (2007). The
New Consort gold mine, Barberton greenstone belt, South
Africa: orogenic gold mineralization in a condensed
metamorphic profile. Mineralium Deposita, 42, 715-735.
Ribeiro-Rodrigues, L.C., Friedrich, G., Meyer, F.M. (1999). The
Cuiab gold deposit, Minas Gerais, Brazil. Erzmetall, 52,
424-437.
Kolb, J., Hellmann, A., Rogers, A., Sindern, S., Vennemann,
T.W., Bttcher, M. E., Meyer, F. M. (2004). The role of a
transcrustal shear zone in orogenic gold mineralization at
the Ajjanahalli mine, Dharwar Craton, South India.
Economic Geology, 99, 743-759.
Kisters, A.F.M., Kolb, J. & Meyer, F.M. (1998). Gold
mineralization in high-grade metamorphic shear zones of
the Renco mine, southern Zimbabwe. Economic Geology,
93, 587-601.
Rogers, A.J., Kolb, J., Meyer, F.M., Vennemann, T. (2013). Two
stages of gold mineralization at Hutti mine, India.
Mineralium Deposita, 48, 99-114.
Billay, A.Y., Kisters, A.F.M., Meyer, F.M., Schneider, J. (1997).
The geology of the Lega Dembi gold deposit, southern
Ethiopia: implications for Pan-African gold exploration.
Mineralium Deposita, 32, 491-504.
Wulff, K., Dziggel, A., Kolb, J., Vennemann, T., Bttcher, M.,
Meyer, F.M. (2010). Origin of mineralizing fluids of the
sediment-hosted Navachab gold mine, Namibia: Constraints
from stable (O, H, C, S) isotopes. Economic Geology, 105,
285-302.
Boer, R.H., Meyer, F,M., L. J. Robb, L.J., Graney, J.R.,
Vennemann, T. W., E. Kesler, E. (1995). Mesothermal-typ
mineralization in the Sabie-Pilgrim's Rest gold Field, South
Africa. Economic Geology, 90, 860-876.
[9] Kisters, A.F.M., Meyer, F.M., Seravkin, I.B., Znamenski,
S.N., Kosarev, A.M. & Ertl, R.G.W. (1999). The geological
setting of lode-gold deposits in the central south Urals: a
review. Geol. Rundschau, 87, 603-616.
Kolb, J., Sindern, S., Kisters, A. F. M., Meyer, F. M., Hoernes,
S., Schneider, J. (2005). Timing of orogenic gold
mineralization at Kochkar in the evolution of the East
Uralian Granite-Gneiss terrane. Mineralium Deposita, 40,
473-491.
18
PROCEEDINGS
Advances in understanding various ore deposits in ultramafic rocks in Indonesia

Ade Kadarusman
Independent Geology Consultant
ABSTRACT
The ultramafic rocks or peridotite are present in the Earths interior as the Upper Mantle constituent beneath the
continental crust or oceanic crust. There exist compositional differences between the lithosphiric mantle underlies the
continent and beneath the oceans. The ultramafic rocks exposed on the Earths surface due to certain tectonic
circumstance. Three principle sources are as (i) orogenic peridotite massifs, (ii) oceanic peridotite; and (iii) peridotitic
xenoliths. Ultramafic rocks exposing in the surface occurred in Indonesia. They are mostly exposed inland of Western part
of Sumatra, Eastern part of Kalimantan, Sulawesi, Halmahera, Banda Arc and Papua. The exposed ultramafic rocks in
Indonesia are mostly derived from peridotite layer of ophiolite rocks; however some of the ultramafic rocks are believe
originally from orogenic peridotite. Ultramafic rocks are the major source of Fe, Cr, PGM, V, Ti, Ni, Co and
Cu deposits. They formed by the magmatic systems as well as weathering processes. The later processes yielded Indonesia
as a major Nickel laterite ores producer in the world in last ten year. Some of deposit may be redistributed by natural
processes, for example platinum placer deposits in South Kalimantan, however until now we do not know the primary
sources for this platinum deposit. Discovery Nickel sulphide deposit in Sulawesi and recently in Sebuku Island South
Kalimantan provided an advance understanding Nickel Sulphide deposit occurrences that may traditionally not exposed in
such tectonic setting in Indonesia. Understanding both tectonic and magmatic systems have a great potential for new
discoveries as occurred recently in the Sebuku Island, this deposits are polymetallic deposit with magnetite
collecting with Ni and Cu sulphides sometime with PGM.
xenoliths, cm to m scale samples carried in rapidly
erupted host volcanics such as alkali basalts and
kimberlites.
I. INTRODUCTION
The ultramafic rocks or peridotite are present in the
Earths interior as the Upper Mantle constituent beneath
the continental crust or the oceanic crust. The poor
knowledge about ultramafic rocks is leading us to
underestimated contribution the ultramafic rock to the
understanding local and regional geological setting, and
their potential to economical ore deposits. In Fact,
Indonesian region, especially Eastern Indonesia is one of
largest in the world of exposed ultramafic rock on the
land. This paper describes the distribution the ultramafic
rocks in Indonesia and their advances in understanding
various ore deposits in ultramafic rocks.
The upper mantle is a peridotitic metamorphic complex
dominated by olivine with increasingly lesser amounts of
orthopyroxene, clinopyroxene and an aluminous phase
(plagioclase, spinel or garnet depending on pressure). The
stability fields of plagioclase, spinel and garnet peridotite
in the upper mantle are dependent on bulk composition.
There exist compositional distinctions between continental
and oceanic crust and so by analogy there is likely to be
compositional differences between the lithosphiric mantle
underlies the continent and that beneath to the oceans.
Orogenic Massifs
These consist of large, several to hundreds of square
kilometer bodies composed dominantly of perodotite with
layers and lenses of pyroxenites. The best localities of the
Alpine type peridotites are found in the Mediterranean
such as Ronda and Beni Bousora. In the Asia, the wellknown example is Haroman Massif in Japan. Dominant
rock types in orogenic massif are lherzolites, with less
than abundant harzburgite and pyroxenite.
Mantle Xenoliths
Xenoliths of upper mantle origin occur in a variety of
host volcanics, of which alkali basalts and ultrapotassic
magma (kimberlites and lamproites) dominate.
Ultrapotassic host erupts prefentially through stable
cratons, whereas alkali basalts generally found in the
younger, tectonically active regions of the continents (e.g
Western USA, Eastern Australia). In the stable cratons,
the ultramafic rocks were sampled by kimberlite-like rock
from the minimum depth of 40 km and later brought up to
the surface as mantle xenoliths. Dominant rock types in
mantle xenoliths are lherzolites, with less than abundant
harzburgite, pyroxenite and eclogite.
GEOLOGICAL SETTING OF THE

ULTRAMAFIC ROCKS
The ultramafic rocks exposed on the Earths surface
(i.e. stable continents, continental margin, island arc or in
the convergent plate boundary) due to the tectonic
circumstance. According to McDonough and Rudnick
(2001), three principle sources are (i) orogenic peridotite
massifs, packages of ultramafic rocks (m to km in scale)
that were tectonically emplaced onto continental crust or
island arc during continental collision, e.g layered
intrusion in the pre-Cambrian rock and Alpine type of
peridotites; (ii) oceanic peridotite, package crustal and
mantle rocks that uplifted onto continental margin or
island arc in ophiolite sequences; and (iii) peridotitic
Oceanic Peridotites
The oceanic peridotite, uplifted in ophiolite sequences
associated with marginal basin onto continental margin,
island arc or subduction complex. In the other words, the
ophiolites are fragment of fossil oceanic lithosphere,
which are found in most mountain ranges of the Earth.
Ophiolite is present because the oceanic lithosphere
escapes from the normal fate (to return to the deep of
mantle, through subduction zone). The ophiolites which
consist of crustal and mantle rocks, and their overlying
sedimentary cover are commonly underlying tectonic
19
PROCEEDINGS
basement and or mlange complexes. The mantle layer of

ophiolite commonly consists of dunite, harzburgite and
lherzolite and lesser amounts pyroxenite and mafic rocks.
Wakabayashi and Dilek (2001) described two types
ophiolite emplacements based on typical convergent
plates boundary occurrences, the Tethyan ophiolite:
collisional or obduction emplacement and the
Cordilleran ophiolite: accretionary emplacement. After
intra oceanic thrusting taken place and subduction
continues, the emplacement of oceanic lithosphere of
Tethys ocean trough obduction is resulted from
continental margin approaches subduction zone, therefore
ophiolite thrust over continental margin, whereas in the
Cordilleran ophiolite occurred within material scrapped
off downgoing plate progressively forms accretionary
complex structurally beneath ophiolite, furthermore
accretionary complex grows with continued subduction.
Ophiolites are largerly classified into two types based
on the differences observed in the geological setting and
geochemistry of several ophiolites studied, i.e supra
subduction zone (SSZ), which exhibits island arc
signatures and Mid Oceanic Ridge ophiolite (MOR). Later
with increasing recognize plume (hot spots)-contaminated
lithosphere, now the ophiolites can be classified as
Oceanic Plateau / Seamounts (OP) origin (Kadarusman et
al, 2004).
More than four huge ophiolite massifs in Eastern

Indonesia have been described petrologically just in last
decade. They are Meratus Complex, East Sulawesi
Ophiolite, East Halmahera Ophiolite, Papuan Ophiolite
and Banda-Arc Ophiolite, respectively. These ophiolites
are very diverse in ages and origins, even in a single
ophiolite massif due to having been generated and
emplaced in differing tectonic settings related to the
extremely complex and protracted triple plate interaction
(Eurasian, Indo-Australian and Pacific). The origins of
these ophiolites ranging from typical mid oceanic ridge to
a supra-subduction zone settings (occur in the continental
margin or island arc as fore-arc basin or back-arc basin),
and recently some of the ophiolites have been proved as
oceanic plateau or oceanic island origin.
UNDERSTANDING ORE DEPOSITS IN
ULTRAMAFIC ROCKS IN INDONESIA
Ultramafic rocks are the major source o f Fe, Cr, PGM,
V,Ti, Ni, Co and Cu deposits. They formed by the
magmatic systems as well as weathering processes. The
weathering-product related deposit such as nickel laterite
or lateritic iron ores are main product of ultramafic rocks
exposed in Earths surface. However, it is not all the
ultramafic rocks can be preserved largerly in the surface,
ultramafic rocks within accretionary complex will not
have significant exposed in the surface and mixed with
many different type rocks other than peridotite, therefore
it will not have significant potential for weathering
process. This weathering-product after ultramafic rocks
yielded Indonesia as a major Nickel laterite ores producer
in the world in last ten year. Some of deposit may be
redistributed by natural processes, for example platinum
placer deposits in South Kalimantan, however until now
we do not know the primary sources for this platinum
deposit.
Sulfides of Cu-Fe-Ni are common accessory phases in
upper mantle peridotites both xenoliths and orogenic
massif occurrences. A large of sulfides compositions have
been reported the ultramafic rocsk association with
magmatism. If the exposed ultramafic rocks were
overprinted by later magmatism that involving underneath
volcanic/granitic basement, then we can expect that base
metal and nickel sulphides deposit (+PGE) is present in
that setting.
Discovery nickel sulphide deposit in Sulawesi and
recently in Sebuku Island South Kalimantan provided an
advance understanding Nickel Sulphide deposit
occurrences that may traditionally not exposed in such
tectonic setting in Indonesia. Understanding both tectonic
and magmatic systems have a great potential for new
discoveries as occurred recently in the Sebuku Island,
this deposits are polymetallic deposit with magnetite
collecting with Ni and Cu sulphides sometime with
PGM. The Cu-Ni-Fe sulphides deposit largely are found
in orogenic peridotite massif, therefore this was proven
that in these island, the ultramafic rocks are not merely
forming as mantle layer or ophiolites. However, the
available worldwide database suggested that SSZ
ophiolite is common for ultramafic rock-hosted chromite
deposits, chromitite and also volcanic-hosted massive
sulfide deposit (Yumul & Balce, 1994). In contrast, mid
ULTRAMAFIC ROCKS OCCURENCES IN

INDONESIA
Ultramafic rocks exposing in the surface occurred in
Eastern Indonesia. They are mostly exposed inland of
Western part of Sumatra, Eastern part of Borneo,
Sulawesi, Halmahera, Banda Arc and Papua. The exposed
ultramafic rocks in Indonesia are mostly derived from
peridotite layer of ophiolite rocks; however some of the
ultramafic rocks are believed originally from orogenic
peridotite as describe in several authors in South
Kalimantan and Central Sulawesi (Monnier et al, 1999;
Kadarusman & Parkinson, 2000). Figure 1 showed
distribution of ophiolite in Eastern Indonesia with the
main constituent of ultramafic rocks exposed in the
surface ultramafic rocks (Kadarusman, 2001). Large
exposed ultramafic rocks on the land are occurred due to
combination of ophiolite obduction or emplacement onto
continental margin/island arc and high rate of crustal
rocks erosion, e.g. East Sulawesi Ophiolite and Halmahera
Ophiolite. Ultramafic rocks within accretionary complex
will not have significant exposed in the surface and mixed
with many different type rocks other than peridotite, e.g
Lok-Ulo complex, Karangsambung, Central Java and
Bantimala complex in South Sulawesi.
Indonesia is only country in the world, which have
huge amount of ophiolite massifs (read: ultramafic rocks)
from two different type of orogenic belt (e.g. Circum
Pacific and Alpine-Himalayan system) scattering
throughout Indonesian archipelago. Ophiolite complexes
of Western and Central Indonesia are believed to have
constituted segments of the Tethys Ocean prior to
accretion along the SE Sundaland margin in the
Cretaceous; those in eastern Indonesia probably represent
parts of the Circum-Pacific Phanerozoic multiple ophiolite
belt.
20
PROCEEDINGS
oceanic ridge origins for ultramafic rocks are almost

barren of economic mineral deposits.
Small scale studies of mineralogy using S EMEDS/QEMSCAN and assay analyses up to ppb
cont ent s using ICP-MS are now have been applied to
study ore deposit in ultramafic in Indonesia, and help us
to understand the processes of formation of these
deposits as well as having the potential to help in the
design of mineral processing of these ores and conduct
beneficiation process of some metal.
REFERENCES
Kadarusman, A., Parkinson, C. D., 2000. Petrology and P-T
evolution of garnet peridotites from central Sulawesi,
Indonesia, Journal of Metamorphic Geology special issue
on Garnet Peridotites and Ultradeep Minerals (editor: J.G.
Liou, D.A. Carswell), v.18, 193-209.
Kadarusman, A., 2001, Geodynamic of Indonesian region; a
petrological Approaches, unpublished PhD Thesis, Tokyo
Institute of Technology, 456p.
Kadarusman, A., Miyashita, S., Maruyama, S. Ishikawa, A.,
Parkinson, C. D. 2004. Petrology, Geochemistry and
Paleogeographic Reconstruction of the East Sulawesi
Ophiolite, Indonesia. Special Issue on Continental Margins
of the Pacific Rim". Editors: Yildrem Dilek, Ron Harris,
Tectonophysic v. 392, 55-83.
McDonough, W.F., Rudnick, R.L., 1998, Mineralogy and
composition of the Upper Mantle, Ultrahigh-Pressure
Mineralogy: Physics and Chemistry of the Earths Deep
Interior, Mineralogical Society of America.
Monnier. C. , M. Polve, J. Girardeau, M. Pubellier, R.C.
Mauryd, H. Bellond, H. Permanaa, 1999, Extensional to
compressive Mesozoic magmatism at the SE Eurasia margin
as recorded from the Meratus ophiolite (SE Borneo,
Indonesia), Geodinnmica Acta (Paris), 12, I, 43-55.
Wakabayashi, J., Dilek, Y., 2001, Emplacement of Ophiolites, In
Ophiolite in Earth History, eds, Y. Dilek and P.T Robinson,
p. 425-431.
Yumul, G.P., Balce, G.R., 1994, Supra-subduction ophiolites as
favorable hosts for chromite, platinum and massive sulfide
deposits.
CONCLUSIONS
The ultramafic rocks or peridotite are present in the
Earths interior as the Upper Mantle constituent beneath
the continental crust or oceanic crust. Three principle
sources of ultramafic rocks exposed on the Earths surface
as (i) orogenic peridotite massifs, ii) oceanic peridotite;
and (iii) peridotitic xenoliths.
The exposed ultramafic rocks in Eastern

Indonesia spread over in Kalimantan, Sulawesi,
Halmahera, Banda arc and Papua. They are
mostly derived from peridotite layer of ophiolite
rocks; however some of the ultramafic rocks are
believe originally from orogenic peridotite that
may produce Cu-Ni-Fe sulphides deposit. In
addition to nickel and cobalt, the ultramafic rock
host for chromite, base metal, platinum group
metals and lateritic iron ores.
21
PROCEEDINGS
Fig. 1 Distribution of Ophiolite in Eastern Indonesia (Kadarusman, 2001)
22
PROCEEDINGS
Characterisation of Coal Deposits for Underground Coal Gasification

Dr. Len Walker
Managing Director, Phoenix Energy Pty. Ltd, Melbourne, Australia
ABSTRACT
Government of Indonesia projections for gas supply and demand suggest that a shortfall in domestic production will require
gas imports to commence in 2021, placing an emphasis on developing other forms of gas production for domestic use.
Underground Coal Gasification (UCG) provides one option for meeting this objective and substantially increasing the
Countrys energy resource base. While the technology has been accepted by the Government as operating under the
current Mining Law, there are a number of technical issues which prevent the application of accepted practices for defining
coal resources and reserves as part of the project approval processes, and as required for financing purposes. This paper
describes the UCG technology and those features which distinguish it from conventional coal mining, and discusses means
by which current resource classification procedures can be adapted to allow the technology to be applied to commercial
project development.
INTRODUCTION
Projections of gas supply and demand made by the
Indonesian Government indicate that a shortfall in
domestic supply is approaching, leading to the country
becoming a net gas importer.
Figure 1 shows that despite the addition of gas supply
from Coal Seam Gas (CSG) production, gas imports will
be required from 2021, and will accelerate over the next
decade. The development of unconventional gas sources
such as shale gas will require substantial exploration
expenditure with significant associated risk.
By comparison, there are substantial coal resources
in Indonesia which are at depths not economic for open
cut mining, and also uneconomic for underground mining,
but which represent a potential domestic energy resource
substantially greater than currently available for open pit
mining.
The process of Underground Coal Gasification
(UCG) is designed to convert this stranded coal in-situ
into a useable gas (syngas) which has the potential to
contribute significantly in minimising the predicted future
domestic gas shortfall.
The commercial introduction of UCG technology
into Indonesia requires resolution of a range of issues,
including selection of suitable coal, development of
economic projects which can support the provision of
conventional financing methods, and the regulatory
approval of the relevant Government Agencies.
Among the approval procedures is the required to
have independent verification of the project coal resource
and reserve size, as defined by one of a number of
accepted procedures, such as the JORC Code (Australia),
NI-43-101 (Canada) or the current Indonesian equivalent.
To apply one or other of these standards to a UCG project
requires an understanding of the process and the issues
involved which differ from conventional coal mining.
Source: Dept. of Energy and Mineral Resources, Indonesia
Fig, 1 Indonesia Gas Supply and Demand
THE UCG PROCESS

In its simplest form, the UCG process is initiated by
drilling two adjacent boreholes into a coal seam, at a depth
greater than 150m (refer Figure 2). A pressurized oxidant
such as air or oxygen/steam is then injected into one of the
boreholes and is ignited at the coal seam. The resulting
chemical reactions within the seam convert the coal to a
syngas, which is then extracted through the second
borehole.
Fig. 2. The UCG Process for Power Generation
The product gas is collected at the surface and treated

to remove residues and contaminants, when it can be used
as either a fuel gas for power generation, or a synthesis
gas for petrochemical processes. Expansion of the process
is achieved by the addition and linkage of further injection
and production wells. Examples of UCG gas production
23
PROCEEDINGS
from the Chinchilla (Walker 2001) and Kingaroy (Walker

2014) projects are shown on Figures 3, 4 and 5.
Fig. 5 Kingaroy (Qld) UCG Gas Flare
Fig. 3 Chinchilla (Qld.) UCG Gas Production
International development of the UCG process has a

long history, which has essentially moved between
countries over the past 50 years. The progress of the
technology can be summarised as follows:
1) Pre 1970s Former Soviet Union
15mt
coal
gasified
2) 1980s/90s U.S.A 30
demonstration
burns
100,000t coal gasified
3) 1990s-2010
Australia, N.Z. 4 pre-project pilot
burns 100,000t gasified
4) Post 2010
Sth. Africa
6 yr. pre-project pilot
burn
est. 50,000t gasified
Fig. 6 UCG in Coal Resource Development

Table 1. Coal Properties for UCG Production Plants
Note 1:
All projects are from the Former Soviet Union,
apart from:
Chinchilla and Kingaroy Australian Projects
Huntly West New Zealand Project
Majuba - South African Project
Alberta Canadian Project (proposed)
Alaska US Project (proposed)
Note 2: 4000kcal/kg = 16.8 Mj/kg
Fig. 4 Kingaroy (Qld) UCG Site Layout
The range of tests undertaken in different countries

and on different coals confirms the conclusion that the
process should be applicable to similar Indonesian coals.
Despite all of the above work, the only location

where a commercial sized UCG plant has operated is at
the Angren coal deposit in Uzbekistan (Gregg, Hill and
Olness, 1976), which in the 1960s produced product gas
sufficient to generate approximately 60MW of power if
used in modern gas turbines. This lack of
commercialisation has been largely due to the withdrawal
24
PROCEEDINGS
of Government support whether by funding or by

political preference.
Typical Coal Seam Properties Required
1) 5m to 15m thick @ 150m to 400m depth
2) Ash + moisture circa 50% max
3) Quality 3000 to 6000 kcal/kg (GAR)
4) No sensitive ground water/aquifers
Establishment of procedures to respond to any

adverse groundwater measurements
Agreement with the Government regulator on
acceptable levels of chemicals in the
groundwater, given the current or likely end use
of aquifer systems
The determination of whether a specific coal seam is

suitable for development thus depends far more on the
characteristics of the integrated coal seam/groundwater
regime than for conventional mining.
The writers experience in reviewing coal deposits
for potential application of the UCG process is that
extensive areas with deep stranded coal deposits exist
which may be suitable for development, both as down-dip
extensions from existing open pit mines, as well as large
unutilised deposits revealed from seismic and oil and gas
well data.
While UCG technology has a clear potential for
creating
new
energy
supplies
in
Indonesia,
commercialisation requires the organisation of a clear
regulatory path to enable development to occur.
COAL DEPOSIT REQUIREMENTS

Figure 6 illustrates a typical coal deposit, with a near
surface outcrop and dip to depths which are typically only
drilled to about 150m. The coal seam may continue to
depths exceeding 300m, depending on the geological
structure of the basin containing the deposit.
The UCG process requires a coal seam at sufficient
depth that, when the coal has been converted to syngas
leaving ash in the remaining cavity, roof collapse will not
extend to the surface to enable gas escape. Development
thus requires a high overburden ratio rather than the low
overburden ratio favoured for open pit mining.
Thus for a preferred coal seam thickness range from
5m to 15m, operating seam depths from 150m to 400m
may be required. Major discontinuities will naturally
affect the process as for conventional underground
mining, as will the geotechnical properties of the
overburden rocks.
In relation to coal quality, previous experience
(especially in the Former Soviet Union) indicates that
calorific values from 3000 to 6000 kcal/kg can be utilised,
with the combined ash plus moisture content not
exceeding 50-55%. The range of coal properties from a
number of locations where the UCG process has
successfully produced syngas is listed in Table 1,
confirming the suitability of the process for Indonesian
coals.
Of equal significance in the development of the UCG
process is the local and regional groundwater regime.
Given that the process generates a range of chemicals in
the operating cavity, it is essential to ensure that these are
converted to volatile products which are recovered in the
production wells, and are not permitted to enter the
groundwater system.
This objective is achieved largely by ensuring that
the pressure of the air or oxygen injection is below that of
the groundwater pressure in the coal seam. This ensures
that any groundwater flow is into the cavity where is
converted to steam to assist in the process. An essential
supplement to this control procedure is a groundwater
monitoring protocol consisting of a number of elements:
Establishment of a base line groundwater
chemical composition in all permeable layers
Installation of a range of groundwater monitoring
bores covering distance from the operating zone,
and depth of the coal seam and other permeable
layers
REGULATORY PROCESSES
UCG has been accepted by the Indonesian
Government as being regulated under Mining Law.
Relevant references are as follows:
Law No.4 of 2009 Mineral and Coal Mining
Defines coal mining efforts to include mining,
processing and purification of coal. Processing and
purification is defined as mining activities to improve
mineral and/or coal quality, as well as to exploit and
obtain any derivative materials
GR 77 of 2014 - Article 94 Sub-article (1):
Coal Value Added
Defines processing or value added by the holder of
Production Operation IUP and IUPK to include coal
gasification including underground coal gasification
Law No. 30 of 2007 on New and Renewable
Energy
Provides for a variety of concessions for New
Energy Sources which includes gasified coal.
With this position established, the procedures
required for approval of a commercial UCG project would
follow those established for conventional mining projects,
viz.
Provision of a Resource and Reserve statement
Submission of AMDAL
Submission of a Feasibility Study
The formats and requirements for the AMDAL and
Feasibility Study follow closely their conventional mining
counterparts, however the preparation of the resources and
reserves statements are not so straightforward due to the
nature of the technology involved.
25
PROCEEDINGS
underground mining project. The question arises as to

their relevance for a UCG project which were clearly not
considered at the time of preparation of the documents.
RESOURCE/RESERVE ASSESSMENT PROCESS

The Australian JORC Code (Ref 4) requires that
Resource and Reserve estimates be undertaken by a
Competent Person, defined by the Code in general terms
as:
5) A Competent Person must have a minimum of five
years relevant experience in the style of
mineralisation or type of deposit under consideration
and in the activity which that person is undertaking.
6) The Code also describes the definition of exploration
results as Mineral Resources or Ore Reserves
using the diagram in Figure 7.
UCG PROJECT SITE CHARACTERISATION

The determination of whether a coal seam is likely to
be developed using the UCG process requires a range of
test data, supplemented by practical experience with the
technology.
Test work will include:
Exploration and in-fill drilling to establish a
potential coal target resource
Determination of coal seam thickness, existence
of multi-seams
Nature and thickness of interburden
Seam depth and overburden ratio
Coal seam properties, including ignition
properties
In-situ gas content (if any)
Coal permeability to air
Overburden and underburden rock properties
Existence of aquifer systems, their permeability
and impact on the process
Groundwater levels, local and regional geohydrology
Groundwater use ie drinking, irrigation etc
Proximity
to
habitation
and
potential
groundwater risks
While much of this work is similar to that required
for conventional underground mining, the nature of the
process itself demands additional relevant data, and more
especially, experience in application of the process to
assess the combined impact of all the information
collected.
Fig.7. Resource/Reserve Definitions
Guidelines to the estimation of a coal Mineral

Resource is contained in an associated document (Ref. 5),
which is to be read in conjunction with the JORC Code.
Any coal Mineral Resource estimate must take into
account the reasonable prospects for eventual economic
extraction. This broad statement implies a general
consideration of the range of factors required for an Ore
Reserve statement, without the necessity for a formal
extraction scoping study.
Of specific relevance to a UCG project coal Mineral
Resource definition, is the description of factors in the
Guidelines (Ref. 5) which should be considered for an
underground mining project: .. aspects such as depth,
faulting, igneous intrusions, working section thickness,
seam dip, physical properties of roof and floor lithologies,
hydrogeology, stress regime, gas content, composition and
permeability should be considered.
In multi-seam
underground deposits, the nature and thickness of the
interburden material may be a critical consideration, as
this might preclude extraction of some of the target coal
seams.
With respect to converting a Mineral Resource
estimate to an Ore Reserve estimate, it is clear that a study
at the Pre- feasibility level is a minimum requirement
Deriving an Ore Reserve without a mine design or
mine plan through a process of factoring of the Mineral
Resource is unacceptable (Ref. 5).
The combination of the JORC Code and the
Guidelines for coal provide a clear path to the estimation
of Mineral Resources and Ore Reserves for an
UCG RESOURCE/RESERVE STATEMENTS

The Australian JORC Code (Ref. 4) and Australian
Coal Guidelines (Ref. 5) provide a sound basis for
Mineral Resource and Ore Reserve statements. However
there is an obvious shortfall in the lack of commercial
UCG operating experience worldwide to provide a
connection between the geological resource estimate and
the reasonable prospects for economic extraction
required for a Mineral Resource estimate, and the
Preliminary Feasibility Study required for the Ore Reserve
estimate.
To the writers knowledge, there is at the time of
writing (October 2016) no small scale UCG demonstration
plant in operation anywhere in the world, let alone being
developed at commercial scale, with the possible
exception being the operation at reduced scale of the
Angren plant in Uzbekistan, the only UCG plant to have
operated at commercial scale some 50 years ago.
The question then arises as to how any Competent
Person assessing a UCG coal deposit can fulfil the
requirements of the JORC (or other) Code to have a
minimum of 5 years experience in the field within which
26
PROCEEDINGS
they are reporting. While this condition is readily met by

many practising geologists in assessing geological factors,
the requirement for an assessment of potential economic
extraction is unlikely to be met.
The JORC Code makes it clear that technical and
economic factors in resource/reserve determinations may
require a team effort in putting together the relevant
assessment. This would require input from a Competent
Person who has been directly involved in UCG
operations.
While a small number of individuals
internationally have been involved in UCG demonstration
plant operation, none has been involved in a commercial
UCG production facility. For UCG technology to become
adopted at commercial scale in Indonesia, this dilemma
must be resolved.
In practice, factors to be considered are:
only a small number of proponents of the
technology in Indonesia are likely to be present
the lack of an existing commercial operation will
limit availability of funding only to small UCG
projects in order to minimise risk
given current understanding of project
economics, a small combined UCG/power
project generating 20-30MW is likely to be
economic (Walker 2014)
such a project will only require identification of a
small coal resource (5-10mt)
UCG project proponents will likely have had
experience of gas production from a pilot scale
UCG project
proponents will be required to provide a
Feasibility Study as part of the permitting process
UCG project proponents will be in competition
with each other, hence will not be in a position to
provide independent advice on either their own
or other projects.
To produce an acceptable Mineral Resource or Ore
Reserve statement, a Competent Person in geological
assessment will have to rely on the project feasibility
study from the proponent to assess potential economic
extraction issues, together with a critical review of other
published information.
Given that early development of the technology will
relate to small projects using the best accessible coal
seams, the likelihood of significant errors in resource
assessment resulting from this process should be limited,
and any resulting consequences minimal.
Expansion of projects beyond the initial small scale
project, requiring a substantial addition to the
resource/reserve base, would then be subject to a further
statement from the Competent Person, who will then have
the existing project operation as a basis for assessing the
economic extraction capability of the expanded project.
UCG POTENTIAL IN INDONESIA

The future commercial development of UCG projects
in Indonesia relies largely on clear guidelines from the
Government in relation to:
Grant of licenses for UCG operations
Relevant laws under which UCG operations are
to be conducted
Processes for approval of Resource/Reserve
Statements, Feasibility Studies and AMDAL
reports
Establishment of technical skills within
Government to monitor UCG operations,
particularly with respect to groundwater
monitoring
Each of these elements is in the process of being
established, and project activity (and associated funding)
must await their completion.
With this framework in place, there is ample scope
for establishment of a very large coal resource base
suitable for application of the UCG process.
The location of all coal basins in Indonesia and their
potential for development of open cut mining is well
documented, as in Figure 8 below.
Fig.8 Indonesian Coal Resources
The compilation of this information is based on

drilling data focussed on shallow coal suitable for open
cut mining, with exploration depths generally limited to
150m maximum.
Little of this coal is likely to be
suitable for the UCG process due to the shallow seam
depth. However geological evaluation of the larger of
these deposits, and the projection of seams down dip, is
likely to identify substantial coal resource targets below
150m as targets for further exploration.
With respect to exploration for suitable coal deposits
meeting the essential properties summarised in Figure 6
and Table 1, existing drilling, sampling and testing
techniques are applicable, other than the requirement that
drilling rigs are suitable to reach depths of say 400m.
Substantial coal deposits at depths suitable for UCG
development can also be identified by use of existing
seismic and oil and gas well data which when correlated
27
PROCEEDINGS
can reveal stranded coal resources which would not be

identified by other means.
The writer has undertaken such a program of work
with the Medco Group and has identified potential coal
resources exceeding 2 billion tonnes in the South Sumatra
Basin alone. While confirmation drilling will be required
to confirm such deposits, there is little doubt the potential
for adding to Indonesias existing energy resource base is
huge. Such a conclusion places more urgency on the need
for putting all Indonesian Government regulatory
procedures in place to enable speedy introduction of UCG
technology.
including clarification of how the relevant code for

defining coal resources and reserves is applied for a UCG
project.
Once the initial project is in operation, there will be
scope for new coal exploration in Indonesias coal basins
to define appropriate deposits to which the technology can
be applied.
REFERENCES
Walker,L.K., Blinderman, M.S. and Brun,K. (2001). An IGCC
Project at Chinchilla, Australia, based on Underground Coal
Gasification (UCG).
Walker, L.K., Underground Coal Gasification Issues in
Commercialisation. Proceedings of the Institution of
Engineers, Energy, November 2014, pp188-195
Gregg, D.W., Hill, R.H., and Olness, D.U., 1976. An Overview
of the Soviet Effort in Underground Gasification of Coal.
Lawrence Livermore Laboratory, University of California.
Prepared for U.S. ERDA under Contract No. W-7405-Eng48
The JORC Code, 2012 Edition. Australasian Institute of Mining
and Metallurgy
Australian Guidelines for the Estimation and Classification of
Coal Resources, 2014 Edition, Coalfields Geology Council
of NSW and the Queensland Resources Council
CONCLUSIONS
The conversion of coal into a useable gas using the
UCG process has a long history in many countries,
although there exists very limited experience in the
development of commercial projects. While introduction
of the technology into Indonesia can substantially increase
the domestic energy resource base, there is much work to
be undertaken before this objective can be achieved.
The immediate emphasis must be on the
establishment of all necessary regulatory procedures,
28
PROCEEDINGS
Successful application of spectral analysis in lithocap exploration and ore deposit

discoveries
Rachel L. Harrison
Independent Consultant Geologist, Joglo Pete, Dusun Pete, Majaksingi, Borobudur, Magelang, Central Java, 56553,
Indonesia. Email: racharrison@gmail.com
ABSTRACT
Spectral analysis by visible near infrared (vis-NIR) and short wavelength infrared (SWIR) spectroscopy, using the
TerraSpec spectrometer has been used successfully in the delineation of world-class porphyry Cu-Au-Mo and highsulfidation epithermal Au-Ag resources at Tumpangpitu in the Tujuh Bukit district of southeast Java, Indonesia. In
conjunction with geochemical, geophysical, regional geological and alteration mapping datasets, spectral data from soil
media has also aided in the discovery of satellite porphyries in the district including the Candrian porphyry at Tujuh Bukit.
Tumpangpitu exhibits a large, complicated example of telescoped Au-Ag (oxide) and Au-Ag-Cu (un-oxidised) highsulfidation mineralization and associated advanced argillic alteration (part of a district-scale lithocap) overprinting the top
of a Cu-Au-Mo porphyry. The Candrian porphyry-high-sulfidation prospect lies 2.2 km east of Tumpangpitu in a remote,
heavily forested region of the project, where surface geological mapping is hampered by limited rock exposures. SWIR
spectral data acquired from soil samples has been used to delineate the surface hydrothermal alteration footprint at
Candrian. Despite the presence of low-temperature clays produced by weathering, careful spectral analysis allowed rapid,
field-based identification of hydrothermal alteration minerals and iron oxides, providing vectors to mineralization in the
lithocap environment (distal, proximal and central to the mineralized zones), and in turn assisted drill targeting. This
technique proved crucial as the porphyry Au-Cu target was concealed under a partially preserved lithocap.
The advanced argillic alteration footprint at Candrian, defined by SWIR analysis of soil samples, extends over an area of
2.5 km by 1.0 km with significant occurrences of pyrophyllite, alunite, dickite, hypogene kaolinite, minor diaspore and topaz
relating to high-sulfidation-style alteration and mineralization. SWIR data identified a white mica and nontronite (Fesmectite) zone; a common alteration product of chlorite, correlating well with hydrothermal magnetite-chlorite porphyrystyle potassic alteration and mineralization intersected by drilling. The spectral alteration data also provided important
insights into the differing erosion levels of the porphyry deposits and prospects at Tujuh Bukit, and therefore, the potential
preservation of high-sulfidation ore in the lithocap.
Iron oxide mapping using vis-NIR spectral data provided important information as a proxy to the distribution of relict
sulfides in the systems. Zones of hematite and mixtures of hematite and goethite are interpreted to reflect oxidation of CuFe sulfides. These zones are coincident with pyrophyllite, high kaolinite and illite crystallinity, Na-rich alunite, annular soil
geochemical molybdenum responses and strong to subtle positive magnetic anomalies in RTP magnetic data, which were all
used to vector towards porphyry mineralization at depth. Concentric halos of goethite that surround the hematite and
mixed hematite-goethite zones are interpreted to replace pyrite in the quartz-sericite halos surrounding the porphyries.
Drillholes at both Candrian were designed specifically based on the results of soil spectral studies in combination with
other supporting datasets. Candrian was drilled in 2011 and successfully intersected porphyry and high-sulfidation style
mineralization. Encouraging drilling results at Candrian include 138 m @ 0.8 g/t Au and 0.21% Cu (CND-11-002) from
just 6 m below surface (Intrepid Mines Ltd. 2011).
The correct auger-assisted soil sampling technique and identification of target soil horizons is absolutely critical for the
success of spectral-aided alteration vector exploration programs. In addition, it is essential to employ experienced spectral
geologists that have the ability to resolve minor signatures of alteration minerals in mixed mineral spectra to be able to
apply the data correctly to produce robust alteration vector maps to aid in ore deposit discoveries.
The application of spectral analysis clearly played a vital role in the identification and delineation of exploration targets
that led to world class discoveries in the Tujuh Bukit district, demonstrating its powerful applicability in epithermal and
porphyry exploration.
x 50 m infill grid conducted by Intrepid in 2011 greatly
increased the resolution of the previously subtle Cu-AuMo anomalies. These geochemical anomalies coincided
with the magnetic highs. After the purchase of a
TerraSpec unit in 2011, the 741 soil samples from this
infill grid were analysed on-site for alteration minerals
and iron oxides. These results, aided in defining and
increasing the dimensions of the alteration system
intersected by the initial six hole diamond drill program in
2011. The results of the spectral study ultimately resulted
in the planning of two additional drillholes, CND-11-006
and CND-11-008, which returned promising results from
intersections of high-sulfidation and porphyry style
mineralization.
INTRODUCTION
Candrian is located 2.2 km east of the Tumpangpitu
high sulfidation Au-Ag-Cu and porphyry Cu-Au-Mo
deposit, part of the Tujuh Bukit project, SE Java,
Indonesia (Fig. 1). A 50 x 50 m infill geochemical soil
sampling survey was conducted by Intrepid Mines
geologists at Candrian from December 2010 through
2011, after a prior regional 200 m x 50 m geochemical
survey in 2009 identified Cu-Au-Mo anomalies to the east
of Tumpangpitu. Magnetic anomalism in RTP geophysical
datasets identified subtle magnetic highs trending
northwest along a structural corridor, possibly reflecting
shallow level hydrothermal magnetite alteration around
porphyry apophyses. The geochemical assays from the 50
29
PROCEEDINGS
SWIR AND VIS-NIR SPECTRAL ANALYSIS

Absorption spectra were obtained with a TerraSpec
model A100810 (350 - 2500 nm) near-infrared and SWIR
spectrometer. The TerraSpec was calibrated with a white
Spectralon disk as a standard to recalibrate the TerraSpec
after every 20 samples. The field of view for analysis was
2cm. Care was taken to analyze representative portions of
the sample matrix, as opposed to unique or uncommon
soil features. Portions of the samples that contained
higher than average concentrations of quartz, sulfides or
magnetite were avoided to minimize reductions in the
spectral response by these opaque minerals. Spectra were
acquired using RS3 software linked to the TerraSpec.
Data processing and spectral analysis was undertaken
manually by the author using The Spectral Geologist
(TSG Pro version 7.1.0.044) processing software
developed by CSIRO for AusSpec International Pty Ltd.
The Vis-NIR was used to determine iron oxide species
using reflectance spectra and SWIR for alteration minerals
using normalised hull quotient spectra for all 741 samples.
A library of absorption features or v-lines were created
in TSG for all alteration minerals including clay, sulfate,
hydroxylated silicate, carbonate and iron oxide species
identified to date at the project. Mineral identification
was supported by comparing each spectrum to a reference
spectral database from TSG and USGS spectral libraies.
This particularly aided the identification of individual
minerals in samples containing up to four or five different
mineral mixtures. TSA (The Spectral Assistant) from the
TSG software was only used as a guide for mineral
interpretation, due to its high degree of inaccuracy when
dealing with mixtures of minerals, as these exhibit a
mixed spectrum that dont match the single mineral
library signatures.
All iron oxides were identified
manually from spectra. Batches of the spectra were also
sent for interpretation by Dr Sasha Pontual of AusSpec
International for QA-QC purposes.
Each spectrum was analysed individually, allowing the
identification of multiple minerals mixtures, as is common
in high-sulfidation epithermal and porphyry systems. For
example mixtures of dickite, pyrophyllite, alunite and
kaolinite were commonly included in one single spectrum.
Each spectrum was analysed in detail using TSA as an
initial guide, and then manually adding a Mineral 3 and
Mineral 4 column to data in TSG software. Alteration
minerals were recorded in order of spectral abundance.
Mineral 1 dominates the spectrum, followed by Mineral 2,
Mineral 3, with Mineral 4 existing only as a minor feature.
This does not necessarily reflect the actual concentration
of alteration minerals in the sample. For example,
minerals such as pyrophyllite and topaz have strong
spectral signatures, even when they exist in small
quantities in mineral mixtures (Pontual et al., 2008).
Fig. 1. Location map of the Candrian prospect, 2.2 km east of

Tumpangpitu, showing the surface alteration signature from
SWIR data over a colored topographic DEM image (modified
from Intrepid Mines Ltd, 2011).
AIMS
1)
2)
3)
4)
5)
The aims of the spectral study on soil samples at

Candrian were fourfold:
To identify alteration minerals including clay, sulfate
and micas from all SWIR spectra and to determine Fe
oxide species from vis-NIR spectra
To calculate and plot spatially spectral parameters
deemed useful for exploration of the prospect,
including illite crystallinity/composition, kaolinite
crystallinity, Fe oxide intensity and alunite
composition
To assist analysis and interpretation of multiple
datasets in delineating vectors to ore (from distal,
proximal and central towards mineralized bodies
To refine and plan additional drill holes to add to the
on-going scout drill program at Candrian.
SAMPLES AND ANALYTICAL METHODS
741 soil samples were taken from Candrian on a 50 m x

50 m grid, marked out by hand held GPS by three teams
comprising a geologist, geotechnical core shed assistants
and local villagers. All team members received thorough
field training on motorised auger sampling from Damien
Lulofs, Chief Consultant Geochemist. During auger
sampling, samples were extracted every 25 cm before
hitting bedrock to ensure that the correct soil horizon was
being sampled. Sampling in alluvial zones was avoided to
ensure geochemical and spectral data were reflecting in
situ, bedrock anomalism. Between ten to fifteen samples
could be taken per day, per team, depending on the terrain
and thickness of vegetation.
Samples of soil for TerraSpec analysis were separated
from the 741 soil samples, placed into RC chip trays,
labeled and air-dried beneath the sun for at least 48 hours
by local, trained geotechnical assistants under the
supervision of the author to allow rapid, independent, data
acquisition.
RESULTS
SWIR spectra from the advanced argillic alteration
zone at Candrian
The surface expression of advanced argillic
hydrothermal alteration at Candrian has been defined over
an area of 2.5 km (northwest-southeast) by 1 km
(northeast-southwest) using SWIR spectral analysis of
alteration minerals in soil samples (Fig. 1). Advanced
30
PROCEEDINGS
argillic alteration at Candrian is characterized by the

presence of Na, Na-K and K-alunite, pyrophyllite,
diaspore, topaz, dickite and hypogene kaolinite.
Alteration minerals identified during analysis of
normalized hull quotient SWIR spectra from Candrian soil
samples, in general order of abundance include:
montmorillonite, poorly crystalline kaolinite PX (the
abbreviation PX defines supergene, poorly crystalline
clay), nontronite (commonly after chlorite), dickite,
crystalline kaolinite WX (the abbreviation WX defines
hypogene, well-crystalline kaolinite), pyrophyllite,
paragonite,
muscovite,
Na-alunite, Na-K-alunite,
diaspore, topaz and minor Mg-chlorite. Almost all soil
samples contained secondary weathering minerals,
whereby poorly crystalline supergene kaolinite (Kaolinite
PX) was dominant. Smectites including montmorillonite
and nontronite were also abundant.
Candrian indicates fluid temperatures in the range of 280 360 C; dickite - bearing samples most likely formed at
lower temperatures (120 - 280 C; Hedenquist et al.,
2000). At Candrian, a clear spatial distribution between
dickite and pyrophyllite has been determined by SWIR
alteration mapping. Dickite occurs in-situ at shallow
levels occupying topographic highs, whereas pyrophyllite
is more abundant down-slope.
This spatial distribution of shallow level, dickitedominant alteration and deeper level pyrophyllitedominant alteration is a characteristic feature at
Tumpangpitu. Pyrophyllite is absent from surface soil and
drillhole data in high-sulfidation state oxide ore Zones A
and C (Harrison, 2011). This suggests that Candrian may
have been subjected to deeper erosion levels than
Tumpangpitu, and therefore may have weaker potential
for the preservation of significant volumes of highsulfidation mineralization.
The advanced argillic
alteration at Candrian is spatially associated with
northwest-trending ledges associated with resistive quartz
- rich bodies in the Candrian lithocap. The advanced
argillic alteration remains open to the west of the Candrian
prospect and may be continuous towards Tumpangpitu. It
may be that the Candrian lithocap amalgamates with the
Tumpangpitu lithocap, which is preserved at higher
elevations. The lithocap at Candrian resembles that of
Tumpangpitu in terms of size, alteration types, zonation
patterns and orientation (Fig. 1) suggesting that it may be
associated with a significant, large mineralized porphyry
deposit similar to Tumpangpitu. Further deep drilling is
required to test this concept.
SWIR spectra from the porphyry alteration zone at

Candrian
Analysis of SWIR spectra is not as useful at identifying
early, high-temperature porphyry-type alteration as it is
with clays associated with advanced argillic alteration.
This is due to the fact that the majority of the key
indicator minerals associated with potassic alteration are
anhydrous (e.g., K-feldspar) or dark (e.g., biotite).
Unfortunately magnetite, K-feldspar, quartz and anhydrite
are all opaque to SWIR analysis. In the porphyry
environment, the combination of visual observations of
alteration assemblages associated with potassic alteration
in drillcore were crucial for developing a final interpretive
alteration map for Candrian. Nonetheless, useful insights
into white mica and smectite, particularly nontronite (Febearing smectite typically forming after chlorite)
distributions in the porphyry sector could be inferred from
SWIR analyses of the Candrian soils.
SWIR spectral analysis porphyry alteration zone

In comparison to advanced argillic alteration which
involves large volumes of hydrous clays (despite large
volumes of associated pyrite; Cooke, 2012), potassic
alteration assemblages contain a larger variety of opaque
minerals such as magnetite, K-feldspar and anhydrite that
cannot be detected by SWIR analysis. However, at
Candrian, a nontronite with minor chlorite zone and a
white mica zone have been delineated from SWIR data,
which are interpreted to be representative of relict potassic
and phyllic alteration respectively.
The spatial
distribution of nontronite, a Fe-bearing smectite that
typically replaces chlorite (Pontual et al., 2008), defines a
northwest-trending corridor to the east of the sampling
grid where drilling has intersected chlorite-magnetiteKspar early, high-temperature porphyry alteration near
surface providing support for this interpretation.
Nontronite zones in the Tujuh Bukit district have
provided an excellent, indirect targeting tool for porphyry
mineralization when correlated with magnetic highs and
soil geochemistry. In comparison to Tumpangpitu, where
the porphyry body is buried at depth below the lithocap,
early high - temperature, and magnetite - rich porphyry
alteration at Candrian is exposed at surface, peaking
through the blanket of advanced argillic alteration.
During exploration in the Tujuh Bukit district, care was
taken to only correlate the presence of nontronite in soil
samples with subtle magnetic highs and associated Cu Au - Mo soil anomalism, and to discriminate them from
nontronite formed after chlorite in the barren propylitic
DISCUSSION
SWIR spectral analysis - mapping the advanced
argillic alteration zone
Significant occurrences of high temperature, acidstable alteration minerals typical of advanced argillic
alteration have been identified. These minerals were
detected in spectra beneath the overwhelming abundance
of weathering-related minerals (characteristic of soil
samples) including poorly crystalline kaolinite and
smectites (mainly montmorillonite).
The high
temperature, acid-stable minerals include alunite,
pyrophyllite, dickite, hypogene, well crystalline kaolinite,
minor diaspore and topaz. Coherent advanced argillic
hydrothermal alteration zones have been defined on
surface maps in Figure from alteration minerals that have
significant abundances (i.e. more than 5%) in the SWIR
soil dataset. They include pyrophyllite, dickite, alunite,
hypogene kaolinite, white mica, diaspore and nontronite dominant zones.
Pyrophyllite alteration occurs within the core of what
are inferred to be roots of the advanced argillic altered
lithocap at Candrian. The persistence of the pyrophyllite
AlOH absorption features, even in samples that contain
only a few percent pyrophyllite, have allowed the
distribution of this alteration type to be mapped
effectively. Abundant pyrophyllite in samples from
31
PROCEEDINGS
zone or elsewhere. Northeast of the sample grid, there is

evidence for chlorite representative of propylitic alteration
bounding the extent of western advanced argillic alteration
associated with Tumpangpitu and Candrian, heading
towards the distal, low - sulfidation Au vein prospect at
Gunung Manis. At Rosebery, Hermann et al, (2001)
found that SWIR spectral analysis proved to be unreliable
for determining chlorite composition in samples that only
contain chlorite in low proportions relative to white mica.
Therefore, although features in the FeOH spectrum were
recognized at Candrian due to the dominance of kaolinite,
smectite and white micas, accurate chlorite compositional
data, could not be obtained.
The composition of white micas (paragonite and
muscovite) were manually identified and calculated from
the position of the AlOH absorption feature. The white
micas generally form a halo around the advanced argillic
alteration zones, with paragonite being more closely
associated with the outer margin of advanced argillic
alteration and muscovite towards the zones of potassic
alteration intersected by drillholes.
The white mica zone defined by SWIR analysis of soil
samples from Candrian may be analogous with porphyry
phyllic alteration, particularly where destruction of
hydrothermal magnetite signatures in RTP datasets is
evident. This alteration type typically displays selective
replacement of potassium feldspar and biotite formed
during early potassic alteration (e.g., Seedorff et al.,
2005). Phyllic alteration is interpreted to be a lower
temperature alteration phase that overprinted potassic
alteration in many porphyry deposits including Pebble
(Harraden et al., 2013), Resolution (Hehnke et al., 2012)
and Lepanto (Chang et al., 2011). At Candrian, the soil
SWIR data shows muscovite occurs at lower elevations
and kaolinite at upper levels, suggesting that these
transition from deep phyllic to overlying advanced argillic
alteration to the west of Candrian.
mineralized or barren hydrothermal alteration that make

the original potassic alteration hard to identify (e.g., at
Tumpangpitu; Harrison, 2011).
The application of
spectral analysis along with visual geological
mapping/logging enables explorers to detect relict early
porphyry alteration and mineralization concealed beneath
lithocap bodies, in particular where rock chip geochemical
exploration techniques have failed to provide any vectors
at surface.
The use of SWIR technology for iron oxide mapping in
surface soil samples has provided important information
as a proxy to the distribution of relict sulfides at Candrian.
The iron oxide intensity spectral parameter provided a
useful guide, but was not as useful as the actual
identification of the iron-oxide phase.
Spectral analysis at Candrian has shown to be useful in
determining vectors to ore in the lithocap environment
both for high - sulfidation state Au - Ag and porphyry Cu
- Au targets. These results should prove applicable
elsewhere, such as when an extensive lithocap exists with
zones of weak to barren mineralization in remote,
inaccessible regions. In such an environment where
geochemical rock chip sampling techniques may fail to
provide direct indicators of the hidden targets, spectral
analysis has the potential to provide an alternative
solution. Identification of concealed porphyry targets at
Candrian under the lithocap, confirmed by the results of
scout drill testing, was facilitated by the integration of
spectral data, subtle soil geochemical data and
geophysical magnetic anomalies which constitute
compelling data for scout drill testing.
Implications for Exploration
Spectral analysis at the Tujuh Bukit Project efficiently
identifies advanced argillic alteration minerals that can be
correlated with geochemical assays, magnetic datasets and
regional geology to provide exploration targets. This
study has illustrated examples of how SWIR spectral
alteration maps can be effectively combined with
magnetic and geochemical datasets to define high sulfidation epithermal and porphyry targets. Subtle,
annular magnetic anomalies, within and along the margins
of a large lithocap, particularly along major structural
corridors require exploration efforts, as the magnetic
anomalism may be caused by hydrothermal magnetite
related to early porphyry-style potassic alteration (e.g.,
Chang et al., 2011). Annular magnetic lows can be caused
by de-magnetization due to quartz - white mica - pyrite
(phyllic) or shallow - level acid clays (advanced argillic)
overprinting potassic alteration, as exhibited at Candrian.
Annual magnetic lows with central discrete magnetic
highs are particularly important as they may be
representative of shallow level magnetite - stable
alteration as shown at Batu Hijau (Hoschke, 2012).
Similar magnetic features have been observed in the
Candrian data, where nontronite can be correlated with
magnetic highs and soil geochemistry to indirectly map
out potassic alteration.
The presence of a lithocap indicates an epithermal
level of erosion, and the potential for epithermal and/or
porphyry mineralization nearby (Chang et al., 2011). This
is clearly demonstrated in the Tujuh Bukit district and at
CONCLUSIONS
Exploration success at Candrian using SWIR data
SWIR spectroscopy on soil samples at Candrian has
proven highly effective at delineating the surface
alteration footprint of the hydrothermal alteration system,
particularly the acid/high temperature - stable
hydrothermal clays of the advanced argillic alteration
assemblage. By using the extendable motorized auger
technique, perfected at the Tujuh Bukit project, soil can be
sampled from greater depths (up to 2 m) adjacent to
altered bedrock. This study has shown that this is a very
effective and low cost technique to create alteration maps
over large forested areas with limited outcrop.
This study has also shown a positive correlation of
early potassic alteration with nontronite after chlorite,
which when correlated with magnetic highs and soil
geochemistry, can be used to vector towards porphyry
mineralization. The identification of nontronite from
Candrian soil data after secondary biotite and shreddy
chlorite and white micas (paragonite, muscovite and
illites) using SWIR spectral analysis are key aspects of
delineating porphyry targets using SWIR data. The
presence of Cu - Au ore - bearing potassic alteration zones
in many cases are masked and obliterated by later weakly
32
PROCEEDINGS
during
hydrothermal
activity,
combined
with
identification of the zones of paleofluid flow through
alteration mineralogy can greatly assist with the
generation of drill targets towards ore. The directional
indicators and mineralization signatures found in this
study have the potential to indicate the likely direction to
the intrusive center during exploration of similar porphyry
- epithermal districts.
many other locations worldwide. Erosion in the Tujuh

Bukit district appears to have been variable in extent since
the formation of the Tumpangpitu lithocap in the late
Miocene (Harrison, unpub. data, 2013). The quartz alunite lithocap at Tumpangpitu is only partly eroded,
whereas it is eroded to deeper levels at Candrian resulting
in porphyry-related features being recognized near surface
at Candrian. Alteration mapping, aided by an onsite
TerraSpec, is essential to assess the advanced argillic
lithocap environment, as is mapping of lithology and
structures. However, alteration by itself may be
insufficient to point to the causative intrusive source.
Mineral parameter vectors found to be effective in this
study include the following:
1) High white mica crystallinity (> 1) adjacent to buried
porphyry bodies
2) Paragonitic white micas (2180 nm - 2190 nm)
adjacent to higher temperature zones of the lithocap
and muscovitic compositions (2200 nm - 2208 nm)
distally
3) High kaolinite crystallinity (> 1) correlating with the
presence of pyrophyllite, indicating proximity to the
high-temperature core of a lithocap and below,
porphyry mineralization
4) Early signs of alunite peak position between 1480 nm
and 1490 nm on the SWIR spectrum shifting towards
Na - alunite at higher wavelengths (>1490 nm) in
samples that are closer to the inferred intrusive source
of acidic condensates
REFERENCES
M., Deyell, C.L., Garcia, J., Gemmell, B.J., McKnight, S.,
and Cuison, L., 2011, Exploration tools for linked porphyry
and epithermal deposits: example from the Mankayan
intrusion-centered Cu-Au district, Luzon, Philippines:
Economic Geology, v. 106, p. 1365-1398.
Cooke, D.R., and Simmons, S.F., 2000, Characteristics and
genesis of epithermal gold deposits: Reviews
in
Economic Geology, v. 13, p. 221-244.
Harraden, C.L., Mcnulty, B.A., Gregory, M.J., and Lang, J.R.,
2013, Shortwave infrared spectral analysis of hydrothermal
alteration associated with the Pebble porphyry copper-goldmolybdenum deposit, Iliamna, Alaska: Economic Geology,
v. 108, p. 483-494.
Harrison, R.L., 2011, The practical application of spectral
analysis in geological modeling andregional exploration at
the Tujuh Bukit project, East Java, Indonesia: Conference
presentation IAGI Makassar, Sulawesi, Indonesia, 32 p.
Harrison, R.L., and Maryono, A., 2012, Tumpangpitu porphyryhigh sulfidation epithermal deposit,
Tujuh
Bukit
Project, Indonesia - geology, alteration and mineralization:
Red Metals Symposium Presentation, CODES, University
of Tasmania, 51 p.
Harrison, R.L., 2014, The application of short wave infrared
(SWIR) spectral data in exploration for the Candrian
porphyry Cu-Au-Mo and high-sulfidation epithermal
prospect, Tujuh Bukit district, southeast Java, Indonesia:
Masters of Economic Geologic thesis, CODES, University
of Tasmania,126 p.
Hedenquist, J.W., Arribas, A.R, and Gonzalez-Urien, E., 2000,
Exploration for epithermal gold deposits: Reviews in
Economic Geology, v.13, p. 245-277.
Hehnke, C., Ballantyne, G., Martin, H., Hart, W., Schwarz, A.,
and Stein, H., 2012, Geology and exploration progress at the
Resolution porphyry Cu-Mo deposit, Arizona: Economic
Geology Special Publication, v. 16, p. 147-166.
Herrmann, W., Blake, M., Doyle, M., Huston, D., Kamprad, J.,
Merry, N., and Pontual, S., 2001, Short wavelength infrared
(SWIR) spectral analysis of hydrothermal alteration zones
associated with base metal sulfide deposits at Rosebery and
Western Tharsis, Tasmania, and Highway-Reward,
Queensland: Economic Geology, v. 96, p. 939-955.
Hoschke, T., 2012, Geophysical signatures of SW Pacific
porphyry deposits: Presentation, Red Metals Symposium,
CODES, University of Tasmania, 8 p.
Intrepid Mines Ltd., 2011, Tujuh Bukit Technical Report, 169 p.
http://intrepidmines.com.au/wpURL
content/uploads/2012/06/
NI43-101-Technical-ReportTujuh-Bukit-Mineral-Resource-PDF.pdf
Intrepid Mines Ltd., 2012, Tujuh Bukit Technical Report, 152 p.
(URL:
http://intrepidmines.com.au/reports-andpresentations/tujuh-bukit-technical-report-november-2012/)
Maryono, A., Setijadji, L.D., Arif, J., Harrison, R.L., and
Soeriaatmadja, E., 2012, Gold, silver and copper
metallogeny of the Eastern Sunda Magmatic Arc Indonesia:
Conference paper MGEI BESA (Indonesian Society of
Economic Geologists, Eastern Sunda Banda Arc) resources
seminar Malang, E. Java, Indonesia, 18 p.
SWIR spectral data, when used in isolation, is unlikely

to produce robust exploration targets. Issues can include
migration of soils downslope, therefore producing an
alteration signature which is not in-situ. To gain the
maximum benefit of this dataset, it has to be compiled
with geochemical, geophysical and regional geology data.
A series of vectors can then be developed that point to the
causative intrusion, which is the potential center to
porphyry mineralization. The gold and copper anomalies
associated with quartz - alunite alteration in a lithocap can
be quite low (<50 ppb), even within a few hundred meters
of the ore zone (Chang et al., 2011; White, 2012; Maryono
et al., 2012). A prospect should therefore not be discarded
based only on a lack of gold and copper anomalism, which
can lead to the erroneous conclusion that it is a barren
lithocap, particularly if the lithocap is large and the
weathering conditions are such that copper can be strongly
leached (e.g., Quimsacocha, Ecuador; Sillitoe, 2010b and
Brambang, Lombok; Maryono, et al., 2012). It is critical
to locate the structural feeders of lithocaps, as high grade
high - sulfidation mineralization, if present, is most likely
to be concentrated there (e.g., Sillitoe, 2010b; Chang et
al., 2011).
In summary, it is crucial to combine field observations
with spectral data, structural mapping, and with
geochemical and geophysical anomalies, when targeting
porphyry and epithermal mineralization. Erosion and
weathering must also be considered, the latter masking ore
in places but potentially improving the ore quality through
oxidation, as is the case at Tumpangpitu. After all of
these datasets are considered, and the SWIR and vis-NIR
data plotted as alteration and Fe oxide species maps,
reconstruction of the topography and hydraulic gradient
33
PROCEEDINGS
Norris, M., 2011, The discovery history of the Tujuh Bukit

copper-gold project, East Java, Indonesia: Conference
presentation, New Gen Gold, Perth, Australia, 70 p.
Pontual, S., Merry, N., and Gamson, P., 2008a, GMEX Spectral
analysis guides for mineral exploration, Volume 1: Spectral
interpretation field manual, Kew, Victoria 3101, Australia,
Ausspec International Pty. Ltd., 189 p.
2008b, GMEX Volume 2: Practical Applications
Handbook: Kew, Victoria 3101, Australia, Ausspec
International Pty. Ltd, p. 51-60.
Rohrlach, B.D., 2011, The discovery history and geology of the
Tujuh Bukit copper-gold project, East Java, Indonesia:
Conference presentation, Round Up, Vancouver, Canada,
35 p.
Seedorff, E., Dilles, J.H., Proffett, J.M., Einaudi, M.T., Zurcher,
L., Stavast, W.J.A., Johnson, D.A., and Barton, M.D., 1995,
Porphyry deposits: Characteristics and origin of hypogene
features: Economic Geology 100th Anniversary Volume p.
251-298.
Sillitoe, R.H., 1995a, Exploration of porphyry copper lithocaps:
Australasian Institute of Mining and Metallurgy, Publication
Series, v. 9, p. 527-532.
1993, Epithermal models: Genetic types, geometrical
controls and shallow features: Geological Association of
Canada Special Paper 40, p. 403-417.
1999, Styles of high-sulphidation gold, silver and copper
mineralization in the porphyry and epithermal
environments: Australian Institute of Mining and
Metallurgy, PacRim 99, Bali, Indonesia, 10-13 October,
Proceedings, p. 29-44.
2000, Gold-rich porphyry deposits: descriptive and genetic
models and their role in exploration discovery: Economic
Geology, v. 13, p. 315-343.
2010, Porphyry copper systems: Economic Geology, v. 105,
p. 3-41.
2010b, Exploration and discovery of base- and preciousmetal deposits in the circum-Pacific region
a 2010 perspective: Resource Geology Special Issue No. 22,
139 p.
Simmons, S.F., White, N.C., and John, D., 2005, Geological
characteristics of epithermal and base metal deposits:
Economic Geology 100th Anniversary Volume, p. 485-522.
Thompson, A.J.B., Hauff, P.L., and Robitaille, A.J., 1999,
Alteration mapping in exploration: Application of shortwave infrared (SWIR) spectroscopy: Society of Economic
Geologists Newsletter 39, p. 1, 16-27.
White, N.C., 1991, High sulfidation epithermal gold deposits:
Characteristics and a model for their origin: Geological
Survey of Japan Report 277, p. 9-20.
ACKNOWLEDGEMENTS
This paper reports selected results from a Masters of
Economic geology thesis completed in 2014 at CODES,
University of Tasmania. Completion of this paper has
been possible with the support, assistance and
encouragement from many individuals and parties. It is
also a milestone in pursuing a passion in gold-copper
deposit exploration and discovery. Therefore, I would
like to express my gratitude to the Tumpangpitu discovery
team members; Adi Maryono, Malcolm Norris, Bruce
Rohrlach, Damien Lulofs, Chris Moore and David Gray
whose strong focus on technical knowledge ultimately led
to the discovery of the Tumpangpitu porphyry deposit in
2008. Special thanks go to my Supervisor Dave Cooke
and all local geologists who contributed to this study
including Mbak Elfina, Pak Rizfan and Paulo Renata.
Additional thanks to logistical support by Mbak Nunung,
Pak Sigit, Mbak Susie and Pak Maruf. Finally thanks to
Intrepid Mines Ltd for initial funding this study.
34
PROCEEDINGS
New Determination of alteration zonations using SWIR in Kyisintaung high

sulfidation copper deposit, Monywa district, Central - Western Myanmar
Yi Sun1 and Jing Chen2
Myanmar Yang Tse Copper Limited (MYTCL), Salingy Township, Sagain Region, Myanmar
ARC Centre of Excellence in Ore Deposits (CODES), University of Tasmania, Private Bag 79, Hobart, Tasmania,
Australia 7001
copper grade is a little higher than the deeper extension
zone.
The predominant syn-mineralization fault is NNEtrending, dipping steeply (65-80) to SE. The mineralized
veins and hydrothermal cemented breccia are controlled
by this group of faults. The vein system occupied the
extension fractures at shallower part of the deposit
(Marjoribanks, 2004). Some sub-vertical (>85) dipping
to SE or NW compression faults with similar NNEtrending observed in deeper part of the deposit also
controlled the vein system. The strike-slip structures are
post-mineralization faults, which only outcropped in post
intrusions (Sun et al., 2015b).
INTRODUCTION
The Monywa copper district is located in the CentralWest of Myanmar, and 115km WWN of Mandalay city.
Over 2 billion tonnes of ore contain 7Mt Cu which are
hosted in this district. There are four high-sulfidation
epithermal deposits in Monywa copper district:
theSabetaung, Sabetaung-South, Kyisintaung deposit, and
Letpadaung 7km to the southeast (Mitchell et al., 2010).
Very limited SWIR analyses have been conducted on less
than 20 drill cores for all of the 4 deposits before (Pontual,
2001; Lazo, et. al., 2006). The results indicated alunite,
pyrophyllite, illite and kaolinite are the predominant
alteration minerals. It also revealed that most pyrophyllite
and illite, unless with moderate to abundant concentration,
were very difficult to identify with petrography studies
(Simpson, et. al., 1996). In this study, we mainly discuss
about the systematically SWIR results on the drill cores of
the ongoing Kyisintaung development project, to examine
the relationships between the alteration and mineralization
types.
METHODS
Half-cut drill core samples were collected from the
ongoing drilling project in the Kyisintaung deposit, with
2-20m intervals based on the variations of alteration
minerals in the drill core logging. Blast hole sludge and pit
rock chips were also sampled to make the cross-validation
during mining.
Shortwave infrared (SWIR) spectroscopy is a
fundamental method to identify the phyllosilicates and
some sulfates. It can help to identify the alteration types in
the porphyry-epithermal systems (Thompson, et al.,
1999). The absorption features observed in SWIR are the
results of the combinations and overtones/harmonics
signals of all the fundamental lattice vibrations in the
mineral molecule which typically occur at longer
wavelengths (Clark et al., 1990; Hunt and Salisbury, 1970;
Hunt et al., 1971).
These samples have been scanned by Portable
Infrared Mineral Scanner (PNIRSTM) with scanning
wavelength ranging from 1300 to 2500nm. The
PNIRSTM instrument was made by Nanjing Institute of
Geology and Mineral Resources and Nanjing Zhongdi
Apparatus Co. Ltd. in China. The spectral result was
interpreted by using Mineral Spectral Analyses system
(MSA) version 3.6 software which was developed by the
same institute. All the data interpretation has been done
manually after the software to prevent any misreading.
GEOLOGY OF THE KYISINTAUNG DEPOSIT

The Kyisintaung deposit lies below a topographic hill
that rises steeply to over 150 m above the surrounding flat
area (Kyaw Win and Kirwin, 1998, Marjoribanks, 2004).
The main host rock is mid-Miocene andesite porphyries,
which intruded the Magyigon formation sandstone.
Pyroclastics including an eruptive diatreme facies
outcropped in southeastern part of the Kyisintaung
deposit. Pyroclastics is separated from the main andesite
porphyries by a NNE tending, steep SE dipping normal
fault zone (Mitchell, 2010. Sun et al., 2015a, 2015b).
A 100-200 m thick barren leached cap situated on the
top of the ore bodies (Sun et al., 2015a). 3 small
supergene-enriched blankets with high clay and sooty
chalcocite only outcropped locally (< 150 m extension in
horizontal and 50-150 m in vertical). The supergene
blankets contain less than 10% ores of the Kyisintaung
deposit. In hypogene area, the ore mineral is dominated by
chalcocite, and minor covellite, digenite and enargite. The
hypogene chalcocite is coarse grained with metallic luster
which can be differed from the supergene chalcocite. The
ores mostly occur as veins, stockworks and cement of
hydrothermal breccias, and steep dipping at deeper part of
the deposit. On the top 100-200m thick of hypogene zone,
minor supergene enrichment can be observed by sooty
chalcocite coating on hypogene minerals and the total
ALTERATION AND MINERALIZATION

Seven main alteration zones were identified based
on the SWIR results and visible geological features:
massive/vuggy silica zone, alunite zone, pyrophyllite
zone, illite zone, sericite zone, dickite zone and kaolinite
zone.
35
PROCEEDINGS
Vuggy silica alteration is not abundant in the

Kyisintaung deposit. Vuggy is notable as veins less than
1m in width. In most cases, vuggy is refilled by silica,
alunite or sulfides. Massive silica is more abundant than
vuggy silica in the Kyisintaung deposit. Nearly all of the
primary texture of the host rock was obscured by massive
silica and associated disseminated pyrite. The pyrite
content in that zone mostly varied from 10% to 30%, but
the total copper grade is mostly lower than 0.2%.
Chalcocite in the massive quartz zone is invisible by bare
eyes.
Alunite is the most abundant hydrothermal alteration

mineral in the Kyisintaung deposit, which overprinted
most other base alterations. Alunite replaces both
groundmass and phenocrysts of the andesite porphyry
pervasively. It occurred in the alteration halos of veins and
hydrothermal breccias. It is characterized by the 1480 nm
absorption peak. This peak location reveals a low
Na/(Na+K) factor of the alunite composition, which
indicate a higher deposition temperature (Chang et al.,
2011). Alunite has a close relationship with the sulfidecemented hydrothermal breccia, which is one of the
dominated copper ores.
Pyrophyllite is one of the main base alteration
mineral and pervasively replaced both phenocrysts and
groundmass of the andesite porphyry. It is characterized
by 2166 nm AlOH absorption peak and 1394 nm -OH
absorption peak. Most pyrophyllite alteration is closely
related with higher silicification. Pyrophyllite is spatially
associated with the overprinted alunite zone. We create an
indicator of the alunite and pyrophyllite as Al/Py, which
is reflected by the depth ratio of the 1478nm and 1394nm
peaks. Al/Py ratio is used to determine the abundance of
alunite and pyrophyllite. In pyrophyllite zone, chalcocite
mainly occurs as solid veins and stockworks with quartz
and alunite halos. Some patchy or open space-fill
chalcocite and minor covellite is also observed in this
zone.
Mica-group is another main base alteration. It
contains 1408-1412 nm -OH peak and 2190-2205 nm
AlOH absorption peak. Several studies have indicated that
lower absorption wavelength of AlOH peak (21902205nm) stands for lower illite crystallinity (Harraden et
al., 2013). So, we divided the mica-group minerals into
illite and sericite subgroups by 2200nm boundary. In both
illite and sericite zones, most copper sulfides are hosted in
solid veins, sheeted veins and minor breccias with massive
sulfide cement. The total copper grade varied and highly
depended on the density of veins.
Dickite is less abundant, mostly occurring as the
crack infill or vuggy infill. Dickite only occurred as the
main base alteration in a small area at the deep centre of
the deposit. Dickite is featured of 1382nm and 1416nm
double absorption peaks. It is spatially and genetically
related
with
covellite.
In
dickite-rich
zone,
covellite/chalcocite ratio is much higher than other area
(>1:1).
Kaolinite is a wide spread alteration mineral which is
related with faults, post intrusions and supergene zones. In
supergene enriched blankets, the kaolinite content ranges
from 15% to 70%. Silicification in kaolinite zone is less
intense than other alteration zone, which decreases
hardness of the host rocks and makes it easy to break.
Chalcocite occurs dominantly as replacing or coating on
the disseminated pyrite in this zone.
IMPLICATIONS FOR PROCESSING (HEAP LEACHING)

Based on the correlation studies of alteration and
copper grade, we found that most high-grade hypogene
mineralization (>1.5% wt. for total copper) is related to
alunite alteration zone. In pyrophyllite, illite and sericite
zone, the copper grade is strongly controlled by the
density of sulfide veins/breccias. Furthermore, different
assemblages and textures of the ore and gangue minerals
in different alteration zone can affect their processing
methods and results.
Heap leaching & SX-EW method is used for
processing due to the good leaching effect of chalcocite.
Covellite is also leachable, but the leaching rate and total
leaching recovery is strictly controlled by the mineral
grain size. During the sulfuric acid and sulfur-oxidizing
bacteria processing, most of the sulfide minerals (pyrite
and chalcocite) will be oxidized and solubilized to the
leaching solution. The clays and micas would partially
change to kaolinite or smectite and the primary texture
will be destroyed. Silica and alunite would not be obvious
affected after leaching.
In MYTCL, leaching pad can be built by crusherconveyer-stacker system or directly by dump trucks. The
different ore transportation methods would lead to
different ore boulder size and compaction rate of leach
pad. High clay ores, if the ore boulder size is too small,
the fine-grained clay minerals will aggregate together
during leaching and significantly reduce the pad
permeability and porosity, and impeding the next layer
leaching. We suggest to transport and tip the high clay
ores directly by dump trucks without crushing, to reduce
the permeability and porosity decreasing effect caused by
clays.
The ore boulder shape or texture of high silica ores
barely changed during leaching which means leaching
solution cannot penetrate the silica surface to inner sulfide
minerals. Considering about this situation, we suggest
crushing the high silica ores, reducing the boulder size to
expose more sulfides to the surface. The covellite-rich
ores should also be crushed to increase the covellite
exposed to surfaces no matter the ore is high silica or not.
36
PROCEEDINGS
Harraden, C.L., McNulty, B.A., Gregory M.J. and Lang, J.R.,

2013, Economic Geology, v. 108, p. 483-494.
Hunt, G.R., and Salisbury, J.W., 1970., Modern Geology, v. 1, p.
283-300.
Hunt, G.R., Salisbury, J.W. and Lenhoff, C.J., 1971, Sulphides
and sulphates : Modern Geology, v.3, p.1-14
Kyaw Win and Kirwin, D., 1998, Proceedings of the Australian
Mineral Foundation Conference, Perth, Australia, p. 61-74.
Lazo, F. B., Maw Oo and Khan Mar Yaw, 2006, Mineralization
and alteration of the Kyisintaung high-sulphidation copper
deposit, Monywa district, Central Myanmar : Report for
Ivanhoe Mines Ltd., unpub.
Marjoribanks, R., 2004, Observations on the geology and
mineralisation of the Kyisintaung copper prospect, Monywa
camp, Myanmar: Report for Ivanhoe Mines Ltd., unpub.
Mitchell, A.H.G., 2010, Geology of the Monywa high
sulphidation copper deposits, Myanmar: Report for Ivanhoe
Mines Ltd., unpub.
Pontual, S., 2001, SWIR spectral analysis of drill core samples
from the Letpaduang high sulphidation system, Myanmar:
Report for Ivanhoe Mines Ltd., unpub.
Simpson, M., Leach, T. and Merchant, R., 1996, Petrographic
and XRD Studies of Drill Core from the Letpadaung,
Kyisintaung, Sabetaung and Sabetaung South, High
Sulfidation Copper Prospects, Myanmar: TLC Report:
96058, Reference: 32604
Sun, Y., Chen, J., Li, X. and Win, Myint, 2015a, The high
sulfidation epithermal copper deposits at Monywa, Central
West Myanmar : SEG 2015 Conference Archives, Hobart,
Australia.
Sun, Y., Peng, H., and Yan, X., 2015b, 2015 yearly review of
Kyisintaung Development Drilling Project : Internal Report
of Myanmar Yang Tse Copper Limited, unpub.
Thompson, A.A.B, Hauff, P.L. and Robitaille, A.J., 1999,
Alteration mapping in exploration: Application of
shortwave infrared (SWIR) spectroscopy: SEG Newsletter,
v. 39, p. 16-27.
Ores in massive silica zone, alunite zone,

pyrophyllite zone and dickite zone, should be sent to
crusher and transported by conveyer. Ores in illite zone,
sericite zone and kaolinite zone should be transported by
dump trucks. In conclusion, alteration information can
help to make better decisions about ore transporting and
processing methods.
CONCLUSION
The Kyisintaung high sulfidation deposit has 7 main
alteration zones whichwere identified by SWIR analyses.
The 3D alteration model shows that alteration is
dominated by pyrophyllite and illite zone. Alunite zone is
steeply dipping with tabular shapes sub-parallel tothe
main vein/fault systems. Massive quartz zone occurred in
the deeper part at the central north of the Kyisntaung
deposit. Dickite zone only occurred as a small stock in the
deep centre. Kaolinite is abundant in the shallow part and
closely related with post intrusions.
Considering the different ore and gangue mineral
assemblages and textures in different alteration zone, the
leaching effect would be distinct. We should use different
heap leaching methods to get better leaching recovery and
leaching rate.
REFERENCES
M., Deyell, C.L., Garcia, Jr., J., Gemmell, J.B., McKnight,
S. and Cuison, A.L., 2011, Economic Geology, v. 106, p.
1365-1398.
Clark, R.N., T.V.V. King, M. Klejwa, G. Swayze, and N. Vergo,
1990, Journal of Geophysics Reviews, v. 95, p. 1265312680.
37
PROCEEDINGS
38
PROCEEDINGS
Ore Pathfinders and Fertility Assessments in Lithocaps: a Case Study of the

Pascua-Lama-Veladero High-Sulfidation Epithermal Au-Ag district, Argentina and
Chile
Lejun Zhang1,2, Noel White1,3, David R. Cooke1,2, Huayong Chen1,4, Francisco Testa1,2, Aldo Vsquez5, Simon
Griffiths5, Jennifer Thompson1,2, and Mike Baker1,2
1
Centre of Excellence in Ore Deposits (CODES), University of Tasmania, Hobart, Tasmania, Australia 7001
Transforming the Mining Value Chain, an ARC Industrial Transformation Research Hub, University of Tasmania,
Private Bag 79. Hobart, Tasmania, 7001, Australia
3
Ore Deposit and Exploration Centre (ODEC), School of Resources and Environmental Engineering, Hefei University of
Technology, Hefei 230009, Anhui, China
4
Key Laboratory of Mineralogy and Metallogeny, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences,
Guangzhou, China 510640
5
Barrick Gold Corporation, Santiago, Chile 222
2
ABSTRACT
The Pascua-Lama-Veladero district is located in the Central Andean Cordillera Frontal of Argentina and Chile, at the
northern end of the El Indio-Pascua Au-Ag-Cu belt, from 3950 to 4450 m above sea level. The results from this study
explains how combining SWIR (short wavelength infrared) and vis-NIR (visible near infrared) data, along with whole rock
geochemical data, alunite, pyrite and quartz trace element chemistry and an understanding of breccia and quartz
paragenesis can be used as effective tools to define HS (High Sulfidation) mineralization targets and indicate the potential
source intrusive centre in a lithocap environment.
and 674.8 Moz Ag (Barrick Gold Corporation, 2013).
Host rocks of Pascua-Lama HS ore are structure
controlled intensely altered Mesozoic granitic rocks and
small stocks of dacite porphyry and granodiorite. The
Veladero deposit hosts 13.8 Moz Au and 226.2 Moz Ag
(Barrick Gold Corporation, 2008). Host rocks of Veladero
HS ore are intensely altered tuffs and polymict breccias of
the Cerro de las Trtolas Formation (16.0 0.2 to 14.9
0.7 Ma) and the Vacas Heladas Formation (12.7 0.9 to
11.0 0.2 Ma; Holley et al, 2012). Two adjacent ore
bodies at Veladero, Filo Federico and Amable, strike
NNW over a combined length of approximately 3 km
(Charchafli et al., 2007).
INTRODUCTION
Porphyry deposits that are not deeply eroded may be
associated with thick and laterally extensive subhorizontal
zones of silicic and advanced argillic alteration named as
lithocaps (Sillitoe, 1995). High sulfidation (HS)
epithermal Au-Ag (Cu) deposits may be hosted in
lithocaps in zones of residual quartz (vuggy quartz,
massive quartz), particularly in their fracture-controlled
roots. The large areal extents of lithocaps, coupled with
the fact that not all lithocaps host HS mineralization,
makes it difficult to predict and define the location of HS
mineralization and to determine whether porphyry-style
mineralization occurs in the underlying intrusive complex.
To enhance and assess the geochemical exploration
tools in lithocap environments, were built by the AMIRA
projects P765, P765A and P1060, Barrick Gold
Corporation provided Pascua-Lama-Veladero district as a
case study site to AMIRA project P1060. The results from
Pascua-Lama-Veladero study site explains how combining
SWIR (short wavelength infrared) and vis-NIR (visible
near infrared) data, along with whole rock geochemical
data, alunite, pyrite and quartz trace element chemistry
and an understanding of breccia and quartz paragenesis
can be used as effective tools to define HS mineralization
targets and indicate the potential source intrusive centre in
a lithocap environment.
ALTERATION AND MINERALIZATION

Widespread zones of hydrothermal alteration have
been recognized in the Pascua-Lama-Veladero district
(Fig. 2). Several stages of hydrothermal alteration are
recognized, including silicic, advanced argillic and argillic
alteration, locally distributed potassic alteration and
phyllic alteration and widespread propylitic alteration
(Williams and Kerkvoort, 2001).
Several styles of Au (-Ag-Cu) mineralization are
recognized at Pascua-Lama and Veladero. The greatest
volume of gold mineralization is contained within an
assemblage of alunite-pyrite-enargite, or occur as native
gold-dominated assemblages (Deyell et al., 2005).
DISTRICT GEOLOGY
The Pascua-Lama-Veladero HS Au-Ag district is
located in the Central Andean Cordillera Frontal of
Argentina and Chile, at the northern end of the El IndioPascua Au-Ag-Cu belt, from 3950 to 4450 m above sea
level (Fig. 1). The Pascua-Lama deposit hosts 17 Moz Au
39
PROCEEDINGS
Alunite SWIR data 1480 nm peak wavelength

increases towards the heat source (Chang et al, 2011). At
Pascua-Lama-Veladero, the higher alunite 1480 nm peak
position values and pyrophyllite, diaspore and zunyite
occur on Penelope and may indicate a hidden intrusive
centre. The same results from alunite vector elements Pb,
Sr, La and element ratios Sr/Pb and La/Pb.
Fig. 4. Spatial plot of alunite 1480 nm peak position values

indicates the heat source is in Penelope.
At Veladero, we only have limited number of alunite

bearing samples. In the alunite bearing samples, several
characteristics still changed systematically with distance
to the Amable HS orebody and the potential intrusive
centre in Veladero deposit including the alunite SWIR
absorption peak at ~1480 nm shifted to higher values
when close to the Amable orebody; Pb concentrations in
alunite decreased toward the potential intrusive centre,
whereas Sr/Pb and La/Pb ratios increased. However
intense and deep weathering complicates alunite
classification and reduces confidence in vectoring based
on the alunite 1480 values and other alunite geochemical
tools. Detailed quartz texture study and quartz trace
elements LA-ICPMS analyses turns out that quartz
chemistry also systematically changed with distance:
elevated Li in quartz indicated the presence of HS
mineralisation in the Amable area, and ratios of Ti/Sb and
Ti/Pb increased toward the potential intrusive centre and
decreased toward the HS ore.
Fig. 1. A) Location of Pascua El-Indio belt. B) Locations of

significant deposits and projects in the El Indio-Pascua belt.
(modified from Charchafli et al., 2007).
CONCLUSIONS
SWIR (short wavelength infrared) data, along with
whole rock multi-elements data and alunite trace element
indicate a hidden intrusive centre located in Penelope at
Pascua-Lama. Quartz can provide vectors to porphyry
centres beneath lithocaps, some elements and ratios
indicate proximity to HS mineralization.
Fig. 2: Alteration map of Pascua-Lama-Penelope-Veladero

district (modified from Barrick Veladero Exploration Team,
2001).
SWIR AND MINERAL CHEMISTRY RESULTS
REFERENCES
To vector to the heat source and assess the fertility of

Pascua-Lama and Veladero deposits, we conducted SWIR
(short wavelength infrared) and vis-NIR (visible near
infrared) analyses, whole rock multi-elements analyses,
and alunite, pyrite and quartz trace element LA-ICPMS
analyses.

M, Deyell, C.L., Garcia, J., Gemmell, J.B., McKnight, S.,
and Cuison, A.L., 2011, Exploration tools for linked
porphyry and epithermal deposits: Example from the
Mankayan intrusion-centered Cu-Au district, Luzon,
Philippines: Economic Geology, v. 106, p. 13651398.
40
PROCEEDINGS
Charchaflie, D., Tosdal, R.M., and Mortensen, J.K., 2007,

Geologic Framework of the Veladero High-Sulfidation
Epithermal Deposit Area, Cordillera Frontal, Argentina:
Economic Geology, v. 102, p. 171192.
Bissig, T., Clark, A.H., Lee, J.K.W., and Hodgson, C.J., 2002,
Miocene landscape evolution and geomorphologic controls
on epithermal processes in the El Indio-Pascua Au-Ag-Cu
belt, Chile and Argentina: Economic Geology, v. 97, p.
971996.
Deyell et al., 2005, Alunite in the Pascua-Lama high-sulfidation
deposits: Constraints on alteration and ore deposition using
stable isotope geochemistry: Economic Geology, v. 100, p.
131148.
Holley, E., Monecke, T., and Bissig, T., 2012, Alunite and
jarosite geochemistry and ages at the Veladero highsulfidation epithermal Au-Ag deposit, Argentina. 2012 SEG
Conference Poster.
Williams, D. K., and Kerkvoort, G. V., 2001. Geology of the
Pascua-Lama project, Chile and Argentina.
ACKNOWLEDGEMENTS
Thanks to Lucio Molina, Federico Wendler and all
who helped us with site access and logistical support from
Barrick Gold Corporation. We also thank all AMIRA
P1060 sponsors and other team members of those projects.
41
PROCEEDINGS
42
PROCEEDINGS
The structure evolution of Zijinshan high sulfidation epithermal deposit, Fujian

Province, China
Jing Chen1, David R. Cooke1, 2, Jos Piquer3, Lejun Zhang1, 2
1
ARC Centre of Excellence in Ore Deposits (CODES), University of Tasmania, Private Bag 79, Hobart, Tasmania,
Australia 7001
2
TMVC (Transforming the Mining Value Chain), Australian Research Council Industrial Transformation Research Hub,
University of Tasmania, Private Bag 79, Hobart, Tasmania, 7001, Australia
3
Instituto de Ciencias de la Tierra, Universidad Austral de Chile, EdificioPugn, Av. Eduardo Morales Miranda,
Valdivia, Chile
ABSTRACT
The Jurassic to early Cretaceous Yanshanian period (180 - 90 Ma) was characterized by major magmatic and
mineralization events in southern China. Subduction of the Pacific plate beneath the Eurasian plate produced large volumes
of volcanic and intrusive rocks. The Zijinshan ore field in southwest Fujian province, southeast China formed at that time,
and is currently one of the largest Cu and Au producers in southern China. The Zijinshan high-sulfidation deposit is located
in the middle of the Zijinshan ore field, and contains 305 t Au and 1.9 Mt Cu. The deposit is hosted in the Zijinshan lithocap,
which developed primarily within the Zijinshan granite complex (157-165 Ma; Jiang et al., 2013). High sulfidation Cu-Au
mineralization is Cretaceous (103 4 Ma; Jiang et al., 2017) and associated with dacite porphyry dykes (105 Ma 2.2 Ma;
Hu et al., 2013).
Keywords: Fujian, Zijinshan, High sulfidation, porphyry
evolution of the Zijinshan deposit have been calculated
using the Multiple Inverse Method (Yamaji, 2000).
INTRODUCTION
The Jurassic to early Cretaceous Yanshanian period
(180 - 90 Ma) was characterized by major magmatic and
mineralization events in southern China. Subduction of
the Pacific plate beneath the Eurasian plate produced large
volumes of volcanic and intrusive rocks. The Zijinshan
ore field in southwest Fujian province, southeast China
formed at that time, and is currently one of the largest Cu
and Au producers in southern China. The Zijinshan highsulfidation deposit is located in the middle of the
Zijinshan ore field, and contains 305 t Au and 1.9 Mt Cu.
The deposit is hosted in the Zijinshan lithocap, which
developed primarily within the Zijinshan granite complex
(157-165 Ma; Jiang et al., 2013). High sulfidation Cu-Au
mineralization is Cretaceous (103 4 Ma; Jiang et al.,
2017) and associated with dacite porphyry dykes (105 Ma
2.2 Ma; Hu et al., 2013).
STRUCTURAL GEOLOGY
The syn-mineralization faults that controlled the
emplacement of dacite porphyry and hydrothermal
breccias are mostly northwest-trending normal faults that
dip moderately to the northeast. Sinistral strike-slip faults
that dip steeply to the southwest were also active during
mineralization, although to a much lesser extent than the
normal faults. The normal and sinistral strike-slip faults
typically have dickite and alunite fault fibres, which
preserve evidence of normal and strike-slip senses of
displacement. Sinistral strike-slip movement post-dated
normal fault movements, based on the horizontal mineral
fibres overprinting the oblique fibres on individual fault
surfaces. The kinematic and dynamic analysis shows a
NNE direction of extension (3) and very high angle
plunges of 1.
Post-mineralization northeast-trending dextral strike
slip faults dip steeply to the northwest. They truncated and
disrupted mineralized veins and breccias. Strike-slip
senses of displacement have been recorded by hematite,
jarosite and minor goethite mineral fibres. This fault
activity relates to WNW-oriented 1 and NE- oriented 3.
DEPOSIT GEOLOGY
Systematic Anaconda-style mapping of the Zijinshan
open pit has been conducted by this study in order to
obtain detailed information about lithotypes, structures,
breccias, veins and alteration mineral assemblages. The
dacite porphyry intruded the Zijinshan granite as dykes,
mostly in the southern part of the deposit. Dykes are either
north-west or east-trending, and steeply dipping.
Mineralized faults controlled the emplacement of dacite
porphyry and tectonic-hydrothermal breccias.
CONCLUSIONS
North-west trending normal faults controlled the
emplacement of the dacite dykes, tectonic-hydrothermal
breccias and mineralized veins at Zijinshan during the
Yanshanian period. The regional-scale fault ShanghangYunxiao fault is NW-trending, located in the south-west
part of the Zijinshan district and was active as a major
normal fault during the late Yanshanian period (Tao and
METHODOLOGY
The dominant fault plane orientation statistics were
analysed by the StereonetTMsoftware (Allmendinger et al.,
2012). The kinematics of fault-slip data was analysed for
variety of lithotypes by the FaultKinTM software
(Allmendinger, 2002). Dynamic analysis of the structural
43
PROCEEDINGS
distribution of the wide range of porphyry Cu-Mo deposits

and epithermal Cu-Au-Ag deposits in SE China (Pirajno
and Bagas, 2002; Qi et al., 2005; Mao et al., 2013). But
more precise geochronology, detailed geological mapping
and geochemistry studies of these deposits need to be
carried out to confirm the geodynamic conditions
prevalent during Cretaceous in this district.
Xu, 1992). NNE-directed extension is consistent with

regional fault movements during mineralization.
Post-mineralization northeast-trending dextral strike
slip faults have been identified throughout the ore field.
The principal compression stress direction was WNW and
the minimum principal stress was NE-trending. This
indicates that the fault activity transferred from an
extensional regime to a strike slip regime soon after
mineralization.
These results are consistent throughout the Zijinshan
district (Piquer et al., 2016). Deformation was widespread
in the Zijinshan district, and was probably related to the
geodynamic evolution of the Pacific margin of SE China
during the Cretaceous. This area transited from
compression to extension from the Jurassic to Cretaceous
and with an extensional regime well established during the
late Cretaceous (Zhou et al., 2006). The precise tectonic
construction of SE China remains controversial. It may be
that the extensional environment was associated with
steeping of the subduction direction of the Paleo-Pacific
plate from oblique to parallel to the continental margin
(Mao et al., 2013). Others have proposed that the trench
retreated, caused slab roll-back and back-arc opening, in
order to create the extensional regime (Charvet et al.,
1994; Zhou et al., 2006; Li and Li, 2007). The first model
is more consistent with the formation and temporal
REFERENCES
Li, Z.-X., and Li. X.-H., 2007, Geology, v. 35, p. 179-182.
Mao, J., Cheng, Y., Chen, M., and Pirajno, F., 2013, Mineralium
Deposita, v. 48, p. 267-294.
Piquer, J., Cooke, D.R., and Chen, J., 2016, Syn-extensional
emplacement of porphyry Cu-Mo and epithermal
mineralisation: the Zijinshan district, SE China: Economic
Geology, submitted.
Pirajno, F. and Bagas, L., 2002, Ore Geology Reviews, v. 20, p.
109-126.
Qi, J.-P, Chen, Y. -J., and Pirajno, F., 2005, Tectonic setting of
epithermal deposits in mainland China. Mineral deposit
research: Meeting the global challenge, v. 1, p. 577-580.
Tao J. -H. and Xu, C. -L., 1992, Geology of Fujian, v. 11, p.
186-203.
Yamaji, A., 2000, Journal of Structural Geology, v. 22, p. 441452.
Zhou, X., Sun, T., Shen, W., Shu, L., and Niu, Y., 2006,
Episodes, v. 29, p. 26-33.
44
PROCEEDINGS
Drones (UAVs) in mining and exploration. An application example: Pit Mapping

and Geological Modelling.
Krisztin Szentpteri1, 2Tatzky Reza Setiawan and 3Arief Ismanto
1
Senior Specialist NBD-GF, PT J Resources Nusantara, Indonesia.

Mine Geologist, Seruyung Mine, PT Sago Prima Pratama, Indonesia.
3
Manager Geophysics, PT J Resources Nusantara, Indonesia.
ABSTRACT
The name drone, for the wider public, generally refers to an Unmanned Aerial Vehicle (UAV) in military applications.
Commercial drones are however light (1-5 kilograms) and small-sized (0.3-1 metres) remote controlled aerial vehicles
made of plastic or carbon fiber and capable of carrying 0.5-10 kilograms payload. They are generally electric motor-driven
and have four- (quadcopter), six- (hexacopter) or eight- (octocopter) propellers or they have the body of a miniature
airplane (fix-winged design). Drones are getting a lot of attention in mainstream media in the past few years and have
numerous commercial applications; most commonly aerial filming and photography, industrial quality control, search-andrescue, security, safety, anti-poaching and trial postal delivery systems.
Commercial drones are also a new toolbox of innovative methods for transforming mining companies as a response of huge
market change in recent years. Typical applications of drones in the mining industry include; aerial surveying, stockpile
management and surveillance of mining installations, leach pads etc.
The application of drone photogrammetry for aerial surveying is also a rapidly developing field in science, natural resource
management and mining. This method was first described by Eisenbeiss 2009 for photogrammetry computer processing of
images acquired by a digital camera attached to a small commercial drone. Since then numerous researchers have used
drone-based photogrammetry for geological mapping of outcrops (Vollgger & Cruden 2016, 2014, Bemis, et. al, 2014,
Vasuki et al., 2014).
However reports on using drones for pit mapping and geological modeling in open-cut mines are still rare, probably
because it is still an ongoing in-house development program for most mining companies.
In this paper we summarize our results using a miniature low cost consumer drone; a quadcopter in open pit mining
environments to create high-resolution pit and bench maps and 3D photo-realistic geological models. The models are
integrated into state-of-the-art 3D geological modeling software platforms. We demonstrate how first order geological
features, on the deposit-scale, can be mapped and digitized into geological models from drone mapping. Such drone models
can supplement typical geological tasks in open pit mining environment such as pit surveying, geotechnical surveying,
geological and structural mapping, bench mapping and various other tasks.
INTRODUCTION
ACQUISTION.
PT J Resources Nusantara, an Indonesian gold miner

and explorer, has recently acquired commercial drones
including three DJI Phantom Professional 3 quadcopters
for an in-house initiative to undertake various tasks.
Although these drone models are designed for film makers
and photographers, they are perfect for the task of aerial
surveying and 3D surface model-building. Our
Exploration Department has commenced a number of
experiments on flying the quadcopters in the various open
pit mining operations of the company. The main task is to
generate 3D models with mapped geological features that
can be transferred to a 3D geological software package for
integration with 3D geology data collected from drill
holes and pit mapping. This paper represents the first
results and findings as well a discussion for further
improvements and of various methods.
Acquiring images using the onboard camera and/or

additional cameras. The DJ Phantom 3 model comes with
a built-in 12Mp camera on a gimbal. The main benefits of
the camera is that it is non-fish-eye and it has a built-in
GPS and altimeter. The images taken will be geo-tagged,
however elevation is measured above the take-off point
not as an Above Sea Level (ASL) measurement. Generally
two camera angles are used; a.) 90 degree downward
looking and b.) 45 degree forward looking (Fig. 1). The
two camera settings are flown the same path one after the
other. A time elapse function allows images to be taken
every 5 seconds. In our setup we have attached an extra,
90 degree downward looking, Garmin Virb Elite camera
which has additional benefits compared to the DJI onboard camera (Fig. 2). The Garmin camera is GPS enabled
and all images are geo-tagged with ASL elevation. The
resolution is bigger (16Mp) and the fish-eye lens (similar
to GoPro cameras) has built-in image transformation
functions, so distortion can be minimized. However,
without ground-control points the precision of the surveys
is as good as the GPS in the camera, which is 1-5 meters.
The additional weight will shorten flight time to only 1-3
minutes but with the time-elapse on the Garmin being
faster (2 second interval) this setup only requires one
flight.
METHODS
The methodology includes three major parts; 1.)
Acquiring images while flying the drone, 2.) Processing of
images and building 3D models with dedicated software
3.) Importing 3D surfaces and imagery into 3D geological
software and digitized geological elements.
45
PROCEEDINGS
metres and keep the level focused on completing the

traverses or grid. There are many ways to fly a survey grid
with the DJ Phantom 3, these include: 1.) manual 2.)
waypoints
3.)
external
autopilot
applications
(DroneDeploy). Each has its benefits and pitfalls. The
manual mode requires the most skill of flying but we
prefer this operation mode due to its flexibility in
maneuvering tight corners in the pit and flying (drifting
while facing the bench face) along benches when
undertaking 3D bench mapping. We emphasize the benefit
of using a quadcopter versus fixed-wing drones when it
comes to detailed bench mapping where the highangle/vertical face has more details than the flat bench,
which are usually covered by a road or debris.
Typical acquisition flight times depend on the size of
the pit but usually take one to three full batteries (3 x 25
minutes). For better precisions (<1m to cm) a fixed
network of reference-points are recommended to be
established before flying (Fig. 1). These points are clearly
marked at minimum 2m in size and DGPS surveyed by
the surveyor team. The fixed ground control points (fgcp)
are usually at an area where mining activity will not
obscure them in future. We also recommend another 3-5
temporary ground control points (tgcp) those are installed
only for the onset of drone survey and include areas of
active mining in example. Different drone platforms allow
different payload and flight times.
Vollgger & Cruden (2016) have pointed out that; larger
and therefore heavier cameras with larger chips would
improve several parameters of the survey and the resulting
model for the better. These larger payload drones are
usually bigger in size and have five to eight rotors,
therefore more expensive and less portable, but they also
have shorter flight time. Those are probably better suited
for mining applications only, however for regional- and
greenfield- exploration purposes, the lighter and more
portable drones are preferred.
Fig.1 Typical flight pattern and camera setting for the drone
aerial survey. With single camera flight path is flown two times,
with double camera it is flown only one time
PROCESSING.
Processing of images and building the 3D surface
model is taken place by using Photogrammetry software.
There are a number of cloud-based and desktop-based
software available on the market with a price tag from
US$3,000-15,000. We are using the Agisoft PhotoScan
(Agisoft, 2015). Typical number of images of a survey
varies between 100 and 600. In our case the Garmin
camera will have nearly two times more images than the
DJI. The benefit of the Garmin images being geo-tagged
with the correct Z value is that it will correct the Z value
of the DJI images when the pictures are aligned. The
resulting model will have a forward- 45 degree and
downward 90 degree looking image pair for nearly each
point (Fig. 1, 2). The software will process these images
into various outputs, most typically into a Digital
Elevation Model (DEM) surface and an orthophoto. The
outputs are in Latitude/Longitude format so before using
the more common UTM based geology models they must
be transformed using Global Mapper or any similar
packages. The outputs are; a UTM-rectified 3D point
cloud format (Fig. 3), surface (Fig. 4), mesh and/or an
image with typical resolution 0.01m/pixel to 1m/pixel
(Fig. 3). The quality/details of image and surface are the
function of processing time that can take 1 hour to 8 hours
Fig. 2 DJI Phantom 3 quadcopter (drone) equipped with an extra

Garmin Vibr Elite GPS camera. Very compact and field-portable
set, all, controller, iPad screen and battery fit in the black
backpack. Note the DJI built-in camera on the front is 45 degree
looking and the Garmin at the back is 90 degree downward
looking.
The flight path is one of the most important aspects of

the photogrammetry modeling process. The quality of the
photogrammetric surface model depends firstly on the
flight parameters (and secondly the size and resolution of
the camera sensor) that is basically the overlap of the
images taken. For the best results, a minimum 50% side
and 50% front- laps are required. The overlaps can be
calculated from the flight and the camera settings
themselves; the altitude, the speed and the time-laps
interval and density (spacing) of lines and/or grid flown.
For a detailed review of typical survey and camera
settings, for super resolution 3D structural modeling of
outcrops, see Vollgger & Cruden 2016.
In our open pit survey workflow, we have a
standardized common take-off point, which is a high point
on the margin of the pit. We take the drone up 80-100
46
PROCEEDINGS
and usually require powerful desktop computers with nonintegrated graphic cards. The most significant
achievement of this processing is a 3D image i.e. a 3D
point cloud with RGB values attached each point. The
denser the cloud the better the spatial resolution. The
importance of this 3D color image resides in the
subsequent 3D modeling (Fig. 3).
Fig. 4 Solid surface, generated by Agisoft PhotoScan. A partial

DTM model of the Penjom open pit mine, Malaysia. Looking
down. Note; rugged surfaces are trees and lake in the bottom of
the pit.
GEOLOGICAL MODELLING.
In our experimental setup we use Leapfrog 3D
geological modeling software to visualize the resulting 3D
surface model and the 3D image. Leapfrog can import the
surface as various mesh or point formats those can be final
or processed in Leapfrog to build the final surface. Then
the image is imported and draped onto the surface, to
create a photo-realistic 3D model (Fig. 5).
Fig. 3 The result of the photogrammetry processing by Agisoft

Photoscanner is 3D point cloud, a 3D RBG pixel cloud, basically
a 3D image. Top (A): a hauling truck displayed as a real RGB
3D image in MeshLab software..Bottom (B): the same file is
displayed in Leapfrog only as single channel (R) image, shades
vary from 0-255 as yellow to red. Not perfect but may be used
for digitizing lines on the actual pixel points.
One of the typical disadvantages of the

Photogrammetry surveying is that; highly reflecting
surfaces like water and very detailed surfaces like
canopies of trees are poorly resolved (Fig. 4) when
compared to more expensive techniques like LIDAR
surveying. Many artifacts are created when processing
these surfaces in PhotoScan and those later will need to be
post-processed by PhotoScan or other open source
mesh/point-cloud-editing software.
Fig.5 Photorealistic 3D model of the Penjom Mine, Malaysia.

High Resolution imagery draped over digital elevation model
(DTM) generated by Agisoft PhotoScan. Note; zone of
oxidation, sediment (grey) and intrusive (cream) lithologies are
clearly visible. (Looking towards NNE)
It is important to mention that the image draping

theoretically introduces a lot of distortion and the resulting
image is not equivalent of the 3D image; which is the
original 3D RGB image point cloud. At this point the
latest version of Leapfrog cannot visualize a point cloud
as colored RGB pixels. It can only display the numeric (0255) of one channel. However we find using only one
channel satisfactory, our choice is the R channel, and
digitize geological features in real 3D clicking-on the
pixels as snap points when drawing a line with the
polyline drawing tool in 3D. Another typical 3D mining
software (Micromine) however can visualize the 3D
points with RGB values and display the point-cloud as a
read image, so it can be an alternative way to digitize 3D
dxf lines on the 3D image pixels and then export. This is
suggested to be applied for detailed bench geology or
47
PROCEEDINGS
geotechnical mapping when these lines (3D dxf) can be

used to generate structural data back into Leapfrog.
structural disks on the surface by using the digitize

structures function in Leapfrog (Fig. 6,9).
However the detail of geology features visible on a
photogrammetry 3D model is very much dependent on the
deposit type. One would conclude that these models are
perhaps the most useful for structurally controlled gold
deposits like orogenic gold and epithermal veins, but we
have found useful application of these models for high
sulphidation deposits as well (Fig. 7.8). In these latter
deposits alteration and/oxidation boundaries strictly
influence the grade and metallurgy of the ore. These
changes are clearly visible in the photogrammetry model
using the most typical visible light digital cameras.
Geological Modelling using Leapfrog 3D

Currently we prefer Leapfrog 3D software because it
is very dynamic and the 3D photogrammetry model, DEM
surface and image (Fig. 4,5) can be easily and quickly
viewed and sliced in any angle together with the 3D point
data (assay, lithology, alteration). This can help finding
correlation between the main features visible on pit
photogrammetry models and the data i.e. faults (Fig. 6)
and alteration boundary (Fig. 7).
intrusiv
Fig. 6. Photorealistic 3D model of the Penjom Mine, Malaysia is

imported to Leapfrog. Major geological features such as; 1)
faults 2) outlines of tonalite sills (light color) and 3) bedding can
be digitized straight from the model into point data or 3D lines.
Looking towards ESE.
Fig. 8 Same view as on Figure 7, but gold grades are shown,

blue >0.2 t/t Au, green >0.5 g/t Au, red >1 g/t Au
DRONES IN EXPLORATION.
It appears that the most obvious application of drone
photogrammetry-based geology modeling is in the mining
environment to brownfield exploration near the mines.
However we emphasize that drone photogrammetry
surveys are equally useful for regional and greenfield
exploration. Regional exploration requires a lot of
planning based on satellite surveillance, so far mostly
using Google Earth. However a quick drone survey over
the area of interest can be generated in less than an hour.
Using those models, access to points of interest, usually
large outcrops or rocky-river banks, can be find quickly
saving time and money. Also in areas, covered by less
vegetation, numerous sub-crops and outcrops can be
mapped instantaneously. The aerial image and derived
topographic contour map to be used as a base map for
geology mapping. Even the topo survey would be better
resolution than the currently public worldwide 30m
SRTM. In Indonesia illegal mining activity is a growing
challenge on many exploration projects. However, the
location of mine shafts can be easily mapped and activities
of illegal miners monitored, any time, using these highly
agile, light weight and portable drone quadcopters like the
Phantom 3.
Fig.7 Photogrammetry 3D model of the Bakan Mine, Osella Pit,

Indonesia is imported to Leapfrog. Major geological features
such as alteration boundaries (yellow and red lines) can be
digitized straight from the model into 3D lines. In this particular
example clay domains (red lines) within the oxide ore can be
mapped out for selective mining by avoiding the clays for heap
leach processing hence photogrammetry model has direct
application to geometallurgy too. Looking towards E.
If we digitize these features on the 3D surface Leapfrog

will create a best fit plane on the 3D lines hence the
geometry and dip/direction of the structure is obtained.
This could be further enhanced by a time-sequence model
i.e. generated every quarter, and the same feature is
digitized from the number of model surfaces. Otherwise if
point data exist from physical pit mapping, the structural
data (plotted as disks in Leapfrog) can be used to
interpolate the surface for the digitized line further
enhancing geological reality and precision. If very
detailed photogrammetry models have been generated like
for most cases in bench mapping, structural measurement
may be obtained directly from the model placing
FUTURE PROSPECTIVE
We believe that application of drones in geological
modeling is a great opportunity in mining and exploration,
and it will gain further attention in the industry soon.
Although professional services for aerial mapping at least,
already exist we prefer developing of our in-house
technical knowledge base. In this way surveying is more
economic and can be made many times in one year.
Commercial drone technology advances so quickly,
that mostly influences the payload and fight time of the
48
PROCEEDINGS
quadcopter and commercial photogrammetry software.

We have also created various surface models for pit and
stockpile survey, which were found to be very useful for
day to day mining operations. Than we could create
numerous file formats that can be directly imported to 3D
geological modeling software for viewing the model
against geological data. Furthermore we have found that
these models, if precision and accuracy is attained, can be
used to digitize points and polylines of many geological
features, faults, bedding, alteration and intrusive lithology.
These features are often large-scale that may not be
apparent in the dataset therefore direct visualization is a
great advantage especially if they have close relationship
to controlling the gold grades.
Accuracy and resolution of photogrammetry models
can be further improved by ground control points and
better cameras. Our final attempt is to create open pit
models in a time sequence i.e. completing the survey
quarterly along with the typical quaternary geology map
update. This can improve 3D accuracy i.e. drawing the
lines in a time-space sequence to generate the surfaces for
faults and bedding in example. This way of geology
modeling would be satisfactory enough without the need
of super-detailed bench mapping for point structural data
extraction i.e. veins and bedding.
drone platforms. This will open the way, very soon, for
quadcopter-based geophysical drones that will be able to
carry gamma spectrometers and magnetometers in
example. Such systems already exist but still in an
experimental stage. Another quickly emerging opportunity
is the quadcopter-based LIDAR which will likely surpass
the recently so popular photogrammetry method, but
currently still highly-priced.
.
Fig. 9 Detailed pit bench mapping project viewed in Agisoft

PhotoScan. Note bedding and fractures are clearly visible and
can be used for geology and/or geotechnical mapping. The
model was generated by flying the drone in three lines parallel
with the bench at various elevations and using different camera
angles. For this mapping very agile quadcopter drones by
manual control are the most preferable.
REFERENCES
Agisoft LLC, 2015. Agisoft Photoscan Professional. URL:
www.agisoft.com.
Eisenbeiss, H., 2009. UAV Photogrammetry. PhD thesis. ETH.
Zurich.
Sean P. Bemis S.P., Micklethwaite, S., Turner D., James M.J., ,
Akciz S., , Thiele S.T. , Bangash H.A., 2014. Ground-based
and UAV-Based photogrammetry: A multi-scale, highresolution mapping tool for structural geology and
paleoseismology. Journal of Structural Geology, Vol. 69,
Dec 2014, p. 163-178.
Vasuki, Y., Holden EJ., Kovesi P., Micklethwaite, S. 2014.
Semi-automatic mapping of geological structures using
UAV-based photogrammetric data: An image analysis
approach. Computers and Geosciences 69, p. 22-32.
Vollgger, S. and Cruden, A.R., 2014. The future of structural
fieldwork - UAV assisted photogrammetry. 2015.
Geophysical Research Abstracts, Vol. 17, EGU2015-373,
2015 EGU General Assembly 2015.
Vollgger, S. and Cruden A.R. 2016. Mapping folds and fractures
in basement and cover rocks using UAV photogrammetry,
Cape Liptrap and Cape Paterson, Victoria, Australia,
Journal of Structural Geology, Vol. 85, April 2016, p. 168187.
3750
11258.
33
Fig.10 Ariel view of stockpile UAV photogrammetry models.

Intrusive-hosted gold stockpile (light-colored) and sedimenthosted gold (darker color) stockpile. Penjom gold mine,
Malaysia. These models can be later used for volumetric
calculations.
CONCLUSION
In our experimental setup we have managed to create
meaningful drone-based photogrammetry 3D models of
open pit mines and exploration areas by using inexpensive
49
PROCEEDINGS
50
PROCEEDINGS
The Power of Geology Model Supported by 3D Geophysical Inversion and

Database System as A Tool to Generate Exploration Target in Mined Exploration
Area: Case Study of COW A Geology Model at PT Freeport Indonesia
Nur Wiwoho and Wahyu Sunyoto1
PT Freeport Indonesia, GeoEngineering Division, Tembagapura 99930, Papua, Indonesia
ABSTRACT
Exploration for deep underground Cu-Au deposit in the Ertsberg mining district is carried out with surface and
underground drillings. More detailed drilling, with an advanced underground exploration program, may be necessary to
establish an accurate interpretation of the deposit, geology mechanism and upgrade the resource estimate to the measured
category for reserve estimation and feasibility assessments. The integrity of the drilling programs in mined area for
geotech, hydrology, grade infill, metallurgy and exploration purposes are important drill data in the development of a
deposit model. The use of block modeling techniques, the application of geophysical inversion for geology interpretation
and the application of gyro downhole survey for accuracy of geology data improve the quality of geology block model.
Significant improvements to the interpretive models can also be achieved by incorporating current geology concept in
intrusion emplacement and integrated 3D magnetic inversion model which is constrained with geology interpretation. The
development of implicit geological modeling technique and support of the well-managed big database contributes to
improve the geology modeling confidence. The 3D magnetic model must be consistent with known geological interpretation
and ore forming processes. The accuracy of drill data using current downhole survey technology is very important to
support this process. These tools are applied to improve the accuracy of the deposit model and the quality of the resource
and reserve estimates in order to optimize capital investments and reduce the development and operating costs of surface
and underground mining projects.
GBTA and DMLZ deposits are part of East Ertsberg Skarn System which are sub vertical deposits hosted at contact
between New Guinea Limestone group and Ertsberg diorite from 3500m down to below 2500m and drilled as exploration
target after COW A geology model was created and reviewed. The interpreted Ertsberg diorite model continue down
depth beyond existing drill holes were supported by 3D geomagnetic inversion model.
An integration of drill data from other departments with various drill purposes and collected in a a same database system
within the GeoEngineering Division is a tool to build a powerful 3D Geology Model
Some methods to be applied include restoration of
horizons and faults of current interpretation to horizontal
position which was conducted by Mid Land Valley
consultant in 2015, approaching tectonic regional concept
to explain fault relationship in geology model boundary
and geophysical geomagnetic inversion conducted by
Mira Geoscience to ensure the confidence level in geology
modeling.
Mira
Geoscience
Advance
Geophysical
Interpretation Centre (AGIC) conducted 3D magnetic
modelling using aeromagnetic survey data over the
COWA area in 2010. The VPmg 3D forward modelling
and inversion software are used for potential field
modelling. These geophysical data are used to validate
and constrain the geology model developed from the drill
hole information. The forward geophysical inversion
modeling is constructed with COW A geology model
constraint to ensure the geophysical interpretations are
consistent with known geological processes and
observations.
INTRODUCTION
The primary source of information is from drill holes,
the accuracy of the interpretation is primarily dependent
on the number and accuracy of drill holes and their spatial
distribution. In the Ertsberg Mining district, deposits are
located at depths between 2500m and 4200m. Delineation
drilling of these deposits is done to achieve the confidence
level in the resource and reserve estimates that is required
for a mine feasibility study and carried out effectively
both by surface and UG drillings. To improve the quality
of drill hole, location from UG is chosen to reduce the
length of drill hole and drill cost. An advance delineation
program is required involving more detailed drilling,
deposit mapping and bulk sampling.
The mine design is created based on the deposit
model and errors in the model can have significant effect
on mine operating and capital costs. Establishing an
accurate deposit model is key to the accuracy of project
evaluation and has to be started since exploration stages
by applying a high quality of down hole surveys. The
application of advance technology in downhole survey is
always updated since the application of maxibor in 1993
to gyro in 2014 and quality control of the data by project
geology in 2012. Quality control of geology data
preparation as data validation of pre geology modeling
increases the confidence level in geology modeling stage.
GEOLOGY MODEL
Geology model process involves three stages: data
collection and data validation, geological interpretation
and geology modeling. Each stage requires a distinct set
of control procedures and quality of results at successive
51
PROCEEDINGS
stages is dependent on the quality of the process in the

previous stages.
Data Collection and Data Validation
Drill data includes: rock type, alteration, structure,
bedding, assay, collar data and down hole survey are
checked and audited to ensure these data are reliable. The
drilling, sampling, sample preparation, assay procedures,
and data base management procedures are setup in SOP
by PTFI for all deposits in the district and keep reviewed
and updated for any changes. Database system is utilized
to collect all drill data information such as collar data,
downhole survey, logging, assay results, SG, RQD, Point
Load and magnetite susceptibility which are prepared by
Exploration and Geo Data Modeling (GDM) and reviewed
and approved by each project geologist of surface mine,
UG mine and exploration under approval system in
geology server. Database management includes drill hole
name system, time limit to review final collecting geology
data process and core image archives are set up in SOP.
Starting in November 2009 a Data Acquisition and
Maintenance check list was developed for the Sarbanes
Oxley (SOX) review for PTFI. This SOX checklist
incorporates a review of the data acquisition tasks
required for the reserve and resource reporting. The SOX
check list has four main components to be checked, they
are: Maintenance of Existing Data Base, Drilling and
Transport of Drill Core Samples, Core Shed Handling,
Logging, and Sample Preparation and Quality Control /
QAQC Procedures. Isis and Isix files of Vulcan drilling
database are prepared weekly by Geo Data Modeling and
put in the server network and those file can be used for
modeling, drill hole design, reporting and review by all
project geologists.
A drilling coordinator which is assigned to arrange
drilling projects for varying purposes such as dewatering,
TDR, pit slope stability, grade infill, metallurgy, special
project and exploration have to be set up to avoid double
drilling target and cost efficiency. A close monitoring to
Pontil drilling contractor performance including
production and quality drillings is reviewed by drilling
coordinator and contract group. A total of 10 UG rigs and
3 surface rigs are operated within COWA district with
total average meter per month is 11,000m.
The Drill Core Data Process flow chart is divided
into three main categories: Data Management, Drill Core
Processing, and QA/QC Controls for the Drilling Data.
These are the processes currently applied in
GeoEngineering Division where each category has
numerous sub-divisions as shown on Figure 1. A number
of processes are completed at the Timika core shed
including: core photos, logging, geotech logging, density
measurements, magnetic susceptibility recording and core
splitting. Split core is sent to SFKK (Sucofindo Kuala
Kencana) for assays with transfer documentation.
Fig. 1 Drilling and Drill Core Data Process Control
Down Hole Survey

Every point location of geology information must
have northing, easting and elevation coordinate used for
modeling. Those points can be collected from surface and
UG mappings and drill holes. The accurate position of the
geology data is very important for many reasons. In
exploration drilling, a potential orebody can be intersected
as expected and in economic modeling will get accurate
grade estimation and get accurate position of rock, fault,
alteration, broken zone etc.
The borehole path is often assumed to follow the dip
and azimuth of the collar in a straight line to the total
depth of the drill hole. Actually there are many factors that
can have a significant impact on the path of the drill hole
due to hard and soft rock boundary or vice versa, layering,
broken zone, strongly fractured, homogeneous rock and
the angle between collar path against strike and dip of the
bedding. A deviation of two degrees may seem minor,
however at 1,000m depth in a vertical hole, the actual path
of the borehole will have a 35m lateral displacement from
the planned path. Borehole deviation should be considered
normal, rather than exception and we need to know the
exact position of the actual drill hole as opposed to the
planned path, it means knowing X,Y and Z spatial
coordinates of the drill hole through downhole surveying.
Fig. 2 An 8.1m Maxibor length compare to 4m length gyro

installation for wireline method and even shorter to 3m length
for conventional method. Differences between maxibor and gyro
results as measured in the same hole VZW-288 are shown in the
plan view.
The effect of poor down hole survey in geology

modeling are inaccurate geology interpretation of
stratigraphy, fault and intrusion contacts in geology
model, assay grade in resource-reserve model and RQD
52
PROCEEDINGS
value in RQD model. The importance of XYZ geology

data have to be maintenance since exploration stage and
the procedure to do down hole surveys have to be
standardized.
The technology in downhole survey keeps updated
recently following the development of geology and
geophysic modeling software and core oriented as well.
The usage of downhole surveys in PTFI also develops
from a conventional method, optical maxibor until current
MEMS sensor gyro successively. Conventional methods
were applied in the past before 1993 using acid tube
which measured inclination only, then was followed by
photo bore and sperry sun which are influenced by
magnetic on azimuth reading. The operator of the tools
also developed from initially operated by Pontil drilling
contractor to currently by the third party of PT ENJ for
independency and quality. All down hole survey results
are stored in database system and those are classified in
terms of confidence levels.
GEOLOGICAL INTERPRETATION
In general geology of the Ertsberg district was
formed by homoclinal north limb of the Mapenduma
Anticline dipping to the north on the south side of the
Ertsberg diorite, changing to tight fold at the centre of
Grasberg Intrusive rock, followed by reverse fault and
strike slip fault, and then intruded by 2 major and smaller
intrusions which varied in ages from 4.4 Ma to 2.5 Ma.
Geology Structure
Regional faults in COWA are used to build fault
block domains. Not all regional faults in COW A model
have fault relationships due to structural geology
complexity. A current geologic interpretation based on
current geology understanding and concept are visualized
in COW A Geology Model. A challenge of current
geology model to be applied for geotechnical application
need further review in geology mechanical process. A
collaboration work with structural geology consultant was
set up to review fault framework, horizon interpretation
and intrusion mechanism.
Regional structures which have NW-SE and NE-SW
orientations in the Ertsberg District which cut across
Miocene age of New Guinea Limestone Group are
products of reactivation of Neoproterozoic basement
faults ( MVE report - 2014) which have similar trend with
seismic data in Money Soul Basin in Arafura sea. The
Mapenduma Fault (Foreland Frontal Fault or Foreland
Thrust Zone) is a deep detachment fault which cut
basement through the Cenozoic rocks. This fault is
required to explain the thick skin tectonics where the
folding from mile 50 area to Darewo Fault Zone on the
north part was driven by this basement fault reverse
movement due to subduction along the Darewo Fault
Zone. The thrusting mechanism of this basement fault is
followed by reactivation of Neoproterozoic basement
faults (NE-SW and NW-SE structures). Possibly
reactivated during extensional collapse after delamination
of the Australian Plate. These basement Neoproterozoic
faults are reactivated as left lateral strike-slip faults during
the Central Range Orogeny (CRO).
Intrusion
Fig. 3 QC graphs to review the consistency of azimuth and

inclination measurements in DZ30-01GT-09 hole starting from
in-progress 1 to 4 and completed one.
The current interpretation of COW A in which

stratigraphic horizons are projected through the intrusive
bodies, implies an emplacement mechanism of
assimilation as the dominant agent rather than opening
pull apart concept. The current intrusion concept suggests
that the primary emplacement of intrusion mechanism is
space being made by left-lateral strike slip faulting (i.e.
Luck 1999; Cloos and Sapiie 2013; Sapiie and Cloos
2013). There are no field evidence that large magma
bodies intrude crust and push apart wall-rocks, but dating
and cooling history of large intrusions show incremental
emplacement (e.g Ertsberg diorite).
Completed and in progress holes have to be surveyed

and each hole is surveyed twice IN and OUT of holes to
see the consistency results using either wire line or
conventional methods. Some challenges in using gyro in
the drill site are water pressure, slow survey process per
run in surface drilling and tight blasting schedules. A
measurement in the collar position is very important and
effort to improve the quality in measuring collar azimuth
and inclination is by using TN14 Gyro Compass.
53
PROCEEDINGS
techniques which have not been used in COW A

Geology Model to investigate key model components
including the fault framework, horizon interpretation and
intrusion emplacement mechanisms.
Fig. 5 The forward model (MVE A4 report 2014) revealed pre

and post intrusions which include steepening and rotation of
horizons between the Wanagon and Ertsberg 3/Idenberg faults.
Geology Modelling
COW A Geology model was built in Vulcan in
July 2002 with dimension 13,500m x 13,500m x 5,100m,
rotated 128 for X axis, block size for parent is 300m x
300m x 300m and subblocking is 15m x 15m x 15m. The
current COW A interpretation is highly uncertain below
1500m level due to no drill data. A total of 64 fault block
domains created based on 25 regional faults and 1 fault
block domain of GIC-Limestone boundary. Each fault
block domain is separated by surface triangulation of
regional faults and produce each one as a solid
triangulation and tight each other. A total of 44 variables
including text values are generated in this geology model
(Figure 6).
Fig. 4 Top: cross section interpreted movement history for the

formation of the Mapenduma anticline. Bottom: Tectonic history
where Mapenduma anticline happened after collisional
delamination (After Weiland and Cloos 1996).
Limited drill hole data in the COW A model below

the 2500 m level and interpreted the Ertsberg diorite
continuation down depth using pure geomagnetic
inversion method suggest that the dimension of intrusion
is getting wider down depth than at shallower level.
Current proposed geology model by MVE indicates that
the intrusion dimension is smaller down depth and it
happened in between Ertsberg Faults no 2 and 3 while the
interpretation from a geomagnetic constrained inversion
model indicates that intrusion was emplaced as a sheet
along steeply-dipping bedding planes and this
interpretation is different compared to previous pure
geomagnetic inversion.
The forward modelling suggested by MVE in 2015 is
being monitored by comparing with new drill data
information and will be applied to update COW A
model if it matches with current drilling and mapping
data. These interpretations are supported by restoration
Fig 6: Fault block domains in COWA
A complicated modeling process is generated to

produce COW A Geology model. There are 3 block
model file (bmf) in model processing to get a final one.
After defined block definition file (bdf) and create block
calculation file (bcf) through LAVA script then the model
processing continued by block model initialized. The
block model initialized means Vulcan machine preparing
blocks for all variables then registering them in each
54
PROCEEDINGS
blocks. The next step is then running block calculation file

(bcf) which is already set up to produce block model file
(bmf).
Variables
Stratigraphy variables consist of 14 formations. A
formation contact is built using string and connected each
other along x-section to generate surface triangulation. A
stratigraphy solid is generated as a product of fault block
domain solid cut by 2 surface triangulations. By applying
64 fault block domains producing 552 stratigraphy solid
triangulations.
Fig. 8: COWA Geology Model
Fig.9 Slice variable in COW A Geology Model
GEOLOGY MODELING APPLICATION

Key issues in the geology model are: collar position
and downhole survey, formation and intrusion contacts,
and fault contact update. COW A Geology model was
created in July 2002 with limited drill data (5,945 holes or
1,525,656m) compared to current condition in May 2016
with 12,628 holes or 3,259,364m. Some improvements of
rock type, alteration, fault, intrusion contact and
stratigraphy have been updated quarterly and used for
further RQD and grade estimation models, exploration
target and other drill targets such as dewatering,
geotechnical, grade infill etc.
DMLZ (Deep Mill Level Zone) is a skarn deposit in
EESS (East Ertsberg Skarn System) which is located
below MLZ BC. Some deep holes from upper level
indicate that Copper mineralization continued down depth
in skarn and altered diorite. A followed up delineation
drilling target is generated to add resource and convert
current resource to reserve based on current geology
model of alteration, intrusion contact and resource
boundaries.
A new concept of intrusion emplacement as
explained above is used to review the shape of intrusion
contact by identify interpreted dilation and contraction
contacts between intrusion and skarn shapes. This
interpretation starts to be developed in COW A for
further structural mechanism study.
Fig. 7 Stratigraphy contact starting from 10m strings, surface

and solids in GRS block domain Regional faults, Intrusion and
alteration variables are prepared in solid triangulations which are
free from opening, self-crossing and inconsistency. A total of 21
intrusion triangulations outside GIC and 30 alteration
triangulations are prepared to build the model. A Leapfrog geo
intrinsic model is used as a tool to support in preparing some of
these triangulations either create a new triangulation or boolean
process.
Slice Variables
Slice variables are provided to show combination of
some variables appearance when a block model is sliced
vertically or horizontally. This sliced variables are named
as FORM and there are 6 FORMS in current COW A
geology model (table 1).
Table 1: slice variables in COW A Geology Model
55
PROCEEDINGS
detail in an existing deposit. Geophysical survey had been

conducted since reconnaissance exploration stage in 1992
along COW A and COW B of central range Papua by
PT Freeport Indonesia. Aeromagnetic survey using fixed
wing and helicopter to measure magnetic data were
conducted by consultant with flight line spacing ranged
between 150m and 400m. Aeromagnetic data have been
modelled by SGC (Southern Geoscience Consultant)
which incorporates topography and magnetic data then
continued by following exploration stages based on
analyzed TMI (Total Magnetic Intensity), RTP ( Reduce
to Pool) and AS (Analytical Signal) image results.
Fig.10 A) Delineation drill target to convert resource to reserve

boundaries and add additional resource in 2005. (Source
Presentation file of The Discovery, Geology, Alteration and
Mineralization of The Deep MLZ Deposit, Papua in IAGI -2005
Indonesian Mineral and Coal Discoveries) B) Interpreted
structural models at EESS ( East Ertsberg Skarn System) using
the shape of intrusion-skarn contact ( Alasdaire Pope, structural
geology workshop April 2016).
Fig.12 TMI (Total Magnetic Intensity) along central range Papua

modelled by SGC ( Southern Geoscience Consultant) in 1992.
In year 2008, Geophysical inversion continued to be

developed where the software incorporates geological
constraints and inverts geophysical data directly for the
geometry of geology boundaries. The new magnetic
inversion technique was applied in COW A to advance
geological understanding of the magnetic characteristics
of the COW A area through geologically-based
inversion.
In July 2010, the Mira Geoscience Advance
Geophysical Interpretation Centre (AGIC) was
commissioned by PT Mineserve to conduct 3D magnetic
modelling of aeromagnetic survey data over the COW
A area, using aeromagnetic data acquired by World
Geoscience Corporation in 1992. In this project, the
VPmg 3D forward modelling and inversion software were
used for potential field modelling using: aeromagnetic
survey flight line data, topography, magnetic
susceptibility from pulp measurement, geology block
model and fault surfaces. Topographic relief was high
(3km) , ranging from about 1km to over 4km elevation.
Proximity to fault was used to weight the magnetic
inversion.
The model inversion stages consist of data
preparation, starting model construction, imposition of
constraints, and assessment of results. The GOCAD
Mining Suite software served as data repository, and
provided the platform for VPmg inversion. The magnetic
susceptibility starting model is geology information
(stratigraphy, intrusion, alteration) which have been
Fig. 11 Actual drills on and miss targets due to drill hole

deflection on section TE10
Proposed drillings that are sometimes not intersecting

the target due to technical drillings, ground condition,
water pressure and other factors can be evaluated by
plotting all actual downhole survey against the target. The
missed target area then can be drilled from other area
which is off section or included in other drilling programs
which is arranged by drilling coordinator. Some factors
that have potential for case above can be anticipated by
reviewing current geology model to get bad ground
condition interval, hard and soft rock boundaries, alternate
rock, massive rock that can be communicated to the driller
in advance.
Geomagnetic 3D Inversion model
Geological goals for geophysical surveys in mineral
exploration may be used to identify potential targets, to
understand the larger scale stratigraphy and structure in
which a deposit might be located, or delineate finer scale
56
PROCEEDINGS
converted to magnetic susceptibility value from geology

block model. The model was exported to VPmg format for
forward modelling where TMI (Total Magnetic Intensity)
is produced from starting model (figure B) and measured
(Figure A). To fix discrepancies this forward modelling
can be imported back into GOCAD as a 3D voxed based
model and update the value according to reconcile results
(figure C) and called as bulk optimization of model
susceptibility.
In Figure 13C, a scaling factor of 4 was determined by
empirically brought the computed magnetic response
amplitudes into reasonably good registration with the
measured amplitudes. Possible explanations for the lower
than expected susceptibility include incorrectly calibrated
susceptibility meter, discrepancy in terms of units or
range, incorrect volume corrections, and magnetic
remanence.
Fig. 13 Comparison of the observed TMI (a) and computed TMI

response (b) of starting model before susceptibility re-scaling.
Computed TMI response of starting model after x4 scaling
applied to susceptibility (c) (Mira Geoscience Pty Ltd 2010
report)
Inversion of magnetic data over COW A,

producing a susceptibility model that fits the aeromagnetic
data to within 18nT RMS. Constrained inversion
techniques were adopted to incorporate geological and
magnetic susceptibility information. The TMI data is
fairly
insensitive
to
the
geometry
of
the
monzodiorite/monzonite intrusive at depth. Therefore, the
intrusive geometry cannot be confidently defined from the
TMI, especially when the deep-seated response is
overprinted by shorter wavelength responses from
shallower magnetic sources.
Ertsberg intrusion model is interpreted getting wider
down depth 3km below the lowest drill depth in 2002 and
extended it down depth using geomagnetic unconstrained
inversion model which in general looks similar
interpretation with constrained inversion model for
interpreted extended Ertsberg intrusion (see figure 14).
Some areas of high magnetic susceptibility from
constrained inversion model seem to be controlled by
mineralization zone in EESS deposit of DOZ and DMLZ
and potential exploration target below DMLZ, but this
inverted magnetic susceptibility model does not indicate
high magnetite susceptibility at GBTA (Gunung Bijih
Timur Atas) which is located vertically at surface level
above DOZ.
Fig. 14 Top: COW A inverted magnetic susceptibility model

2010 by MIRA Geoscience showing high magnetite
susceptibility in cross section. Bottom: COW A geology
model control on high magnetite susceptibility as shown on the
left picture.
In 2002, Freeport also conducted geomagnetic

inversion model through geophysic consultant and the
results indicated that no signature of magnetic anomaly on
Kucing Liar skarn which is adjacent to Grasberg Porphyry
deposit (Figure 15-A). This phenomenon is probably due
to the fact that the Kucing Liar deposits at depth is
covered by strong magnetic signal from Grasberg deposit
at the surface and is therefore hard to identify. By using
constrained geomagnetic inversion against COW A
geology model in 2010, the magnetic anomaly can be
identified better than the previous model (Figure 15-B).
57
PROCEEDINGS
exploration stages by applying a high quality of down hole

surveys. The effect of poor down hole survey in geology
modeling are inaccurate geology interpretation of
stratigraphy, fault and intrusion contacts in geology model
assay grade in resource-reserve model and RQD value in
RQD model.
A preliminary study of constrained geomagnetic
inversion techniques to incorporate geological and
magnetic susceptibility information gives an advance
geological understanding of the magnetic characteristics
of the COWA area. A further study will give a better
result using good pulp measurement data and defining
geology alteration using magnetic properties from drill
hole.
Geology model in a mined exploration area is a
compilation of geology interpretations that have been
standardized to support mine design and exploration
target. Standardized cross section, color legend for
geology information, fault name and other mine design
issues are set up and communicated well to avoid
inconsistency and various geology interpretations. Open
geology minded of new geology concept in geology
mechanism is important which could potentially impact in
modelling, exploration, syn and post mineral controls are
kept maintained to face bigger challenges which need a
better understanding of geology knowledge.
REFERENCES:
Cloos, M., Sapiie, B., Quarles van Ufford, A., Weiland, R.J.,
Warren, P.Q. and McMahon, T.P. 2005. Collisional
delamination in New Guinea: The geotectonics of
subducting slab breakoff. Geological Society of America
Special Papers, 400, 151.
Mira, Geoscience Pty Ltd, 2010, Geologically constrained
magnetic inversion for COWA Indonesia, Consultant
Report to PT Mineserve International, Brisbane QLDAustralia.
Pope, A, 2016, Grasberg Ertsberg structural geology review, in
Structural geology workshop, Tembagapura-Papua.
Pringgoprawiro, A, 2016, personal discussion about geomagnetic
inversion in COWA.
Reflex, 2015, A Guide to borehole deviation and surveying
version 2, Perth-WA.
Independent Mining Consultant, Inc, 2016, Review and
verification of reserves prepared for PT Freeport Indonesia
Company, January 1, 2016.
Valley, M Exploration Ltd, 2014, Report of phase A1:
Validation of a regional-scale cross section for Freeport
McMoran, Glasgow UK.
Valley, M Exploration Ltd, 2015, Report of phase A4: Analysis
of 3D geological model for Freeport McMoran, Glasgow
UK.
Wiwoho, N. and Hughes, S., 2005. The discovery, geology,
alteration and mineralization of the Deep MLZ DepositPapua, in IAGI, Bogor.
Wiwoho, N., 2016, COWA geology model, QTR Report, PT
Freeport Indonesia, Tembagapura.
Fig. 15 Comparison among pure geomagnetic inversion (2002)

Top (A), constrained geomagnetic inversion center (B) and
geology model bottom (C). No drill hole data below 2000m
level and the Ertsberg diorite is extended down in current
geology model using interpreted pure geomagnetic inversion.
CONCLUSION
Establishing an accurate deposit model is key to the
accuracy of project evaluation and has to be started since
58
PROCEEDINGS
59
PROCEEDINGS
Fuzzy Logic and Weight of Evidence Approach for Mineral Discovery Tool
Harman Setyadi1, Budi Santoso, STJ2
1
Doctorate Student, Mine Engineering Study Program Faculty of Mine and Petroleum Engineering, Institute
Technology Bandung, Jl. Ganesa 10, Bandung 40132 INDONESIA. e-mail: Harman.Setyadi@yahoo.co.id;
Harman_Setyadi@jresources.com
2
J. Resources Nusantara, Equity Tower, 48th floor, SCBD. Jl. Jendral Sudirman Kav 52-53 Jakarta 12190.
ABSTRACT
Mineral Exploration is a high risk and uncertain business. To reduce the risk, exploration should be done carefully and
thoughtfully step by step. Decision should be thoroughly assessed properly, not proper decision due to lack of proper data
analysis will lost the opportunity to discover a new economic ore deposit. Fuzzy logic and weight of evidence (WofE) was
implemented widely on the industry such as for a controller and predictive tool to improve the decision making. This method
also was used widely for mineral exploration prospecting tool, however it is practicaly not simple due to the data
processing complexity and software specific requirement.
This study was proposed to implement simplified fuzzy logic and WofE for alteration mapping and mineral prosperity
modelling, using the geophysical data. Fuzzy logic is the method to simplify the exploration data classification included the
anomaly level determination. WofE is the probabilistic method using the Bayesian roles, which is widely used for the
predictive modelling. The combination of the Fuzzy logic and WofE is take the light table geologist prospecting method
in the past. By implement this method mineral discovery should be more effective by reducing the subjectivity assessment
and able to cover all area quickly.
Key words: geophysical; fuzzy logic; weight of evidence (WofE); Mineral Discovery
other information such as geophysics and geochemistry
data are available.
Another issues in the exploration decision making is
the data readiness. Most of the (previous) exploration data
are not well stored in a standard and proper format and
centralized location. When exploration team does the drill
target delineation and/or run a drilling program,
sometimes they make decision by using partly ready data
only. The presence of an adequately stored and proper
existing data sets are critical for the identification of
prospective ground (Scott and Dimitrakopoulos, 2001).
Accordingly, the data management with assured quality is
very important and is one of keys of success in
exploration (Setyadi, 2012).
Discovery of new mineral resources is not an easy
task and yet the probability is quite low. The challenge of
current and future mineral exploration is to discover new
economic mineral deposits which are not well exposed
(deep) and only have weak anomaly signatures (Setyadi,
2013a, 2013b). Beside high cost, it requires more data
with better technique to process and interpret combined
with better understanding in geology conceptual model.
Exploration geophysics and geochemistry have been
considered to be very powerful and proven mineral
exploration tools, they usually generate large data set.
Proper quantitative data analysis and evaluation effort are
required to generate viable exploration targets in both
regional and prospect level in exploration stage. The
power of utilization large dataset and implementation of
some techniques in regional scale exploration work have
been done by several parties/companies and They have
proven to be very successful in delineating exploration
target, but for smaller scale such as prospect scale is
INTRODUCTION
Mineral ore body is a dynamic entity which is prone
to fluctuation of the metal price. The current situation with
the declining metal (gold in particular) price has severely
affected the (metal) gold exploration and mining business
sector. Computer with special designated software should
be helpful to recalculate/re-simulate the economic
parameters/calculation (Sinclair, and Blackwell, 2004).
Computer; with special software and specific yet creative
geological approach will also help to make better decision
in the exploration activity, such as delineating, ranking
drill target and optimizing drilling program.
The depletion of ore mineral in the world has become
a very big issue, exploration is the only the way to
replenish but those near surface deposits have probably
been mostly discovered whereas to explore and discover
deeper ore mineral deposit is not easy and very
challenging and costly. One of the reason the decline of
mineral exploration discovery is thought due to the
exploration decision that is taken only based on the
qualitative interpretation based on partial information
rather than quantitative analysis (Barnett and William,
2012). It is particularly dealing with the fact that the most
if not all of near surface and easy to find deposits have
already been found. Often exploration regional
geochemistry program missed the target and drilling
program delineates unsuccessfully the sub-surface/deep
ore deposit that is commonly due to similar
aforementioned approach. Field geologist tends only to
use partial information combined with the very subjective
and limited geological knowledge to delineate the possible
mineral deposit continuation and/or location, although
60
PROCEEDINGS
Fuzzy logic is one of the other modelling method

which is widely used (Figure 1). Fuzzy logic was firstly
introduced by Zadeh (1965) as mathematical formula to
represent the ambiguity of data. Fuzzy logic converts
variables from the raw data to the probabilistic variables,
based on the estimation rather than exact calculation
which has variable value from 0 to 1, where 0 is non
membership and 1 is full membership (An, Ping., at all,
1991). However the data classification should be done
using rule of statistical calculation. The benefit of the
mathematical approaches are: (a) maximizing data
extraction; (b) effectively combining diverse information;
(c) providing tools to quantify inherent uncertainties; (d)
ranking potential targets; and (e) reducing data processing
and evaluation time (Luo and Dimitrakopoulos, 2003).
something we like to emphasize in this paper with proven

case study.
GIS MODELLING FOR MINERAL PROSPERITY
GIS mineral prosperity modelling was offered by
several researcher with several different mathematical
aproach by processing all available exploration data set
(Bonham-Carter, 1996; Bonham-Carter, Agterberg, 1988;
1989; Carranza, 2004; 2008; 2009; 2010). Commonly at
this time GIS mineral prosperity modelling was done in
the regional scale. Ore mineral modelling was performed
by compiling and integrating overlaying several
exploration maps which each represents respective criteria
and feature of the mineral deposit model (Luo and
Dimitrakopoulos, 2002; Carranza, et. all, 2008). One of
the common method used for modelling is the Weight of
Evidence (WofE) in combination with other method.
Figure 1 is present the statistical research during last three
year. WofE method commonly requires input of converted
map on the binary or ternary maps. Those maps are then
compared and integrated in one map in which we could
present and predict the mineral deposit potential
(Bonham-Carter, 1989; Bonham-Carter, 1996; Cheng and
Agterberg 1999). The prosperity map was created based
on (a) positive value estimation of positive correlation
between prospect and spatial evidence; (b) number of
prospect appear on the spatial zone; and (c) spatial
positive correlation between geological interpretation and
prospect (Carranza, 2004).
METHODOLOGY
GIS prospecting for mineral discovery has same
logical approach of the map analysis over the light table
by geologist in the past. Map overlaying is the map
integration method to figure where the potential of
mineral deposit based on the existing data and geological
knowledge. Geologist will assessing the anomaly level
and boundary, assessing the favourable geological data
such as structure, lithology and alteration. Geologist
should be identify the known mineral deposit both from
mapping or drilling, use the similar anomalies feature to
see the mineral occurrence extent or to search the similar
mineral occurrence which is not discover. On the virgin
area which has limited known mineral occurrence,
geologist will use their knowledge and experience to
discover new deposit based on the combination of
exploration data (map) what they have.
Fig.2
Illustration of prosperity mineral modelling based on

the geological exploration approach (Setyadi, 2016)
Figure 2 present the illustration of the geological

logic how geologist to predict the subsurface ore using the
surface exploration data based on the outcropped ore
body. This logic was used for the weight of evidence
modelling. The outcropped rock was used as the data
training. Data training is one of the important rules for the
data driven method. Data training was created from the
Fig. 1: Publication distribution of the mineral prosperity

modelling (Setyadi, 2016)
61
PROCEEDINGS
selected area which was proper mapped and defined the

specific alteration type and has distinguished geophysical
anomaly.
Table 1: Comparison Conventional and GIS Prospecting Approach (Setyadi, 2016)

Step
1A
Conventional Light Table Map Analysis

Anomaly level assessment of geochemical, geophysical data
GIS Analysis approach

Fuzzy Logic, Fuzzy membership
Favourable Geological features assessment
Fuzzy Logic, Fuzzy membership
Identify what the anomaly level related to the deposit
Posterior probability
Compare and integrated anomaly map to get the anomaly model
Weight of Evidence
Store the criteria and model on his brain and/or paper. Some time is
bias and subjective.
Store the criteria and model on memory,

objective.
Calculate the combination of anomaly occurrence (AND, OR, SUM,

PRODUCT), approximated only
Fuzzy Operator (AND, OR,

PRODUCT), accurate and detail
End Product as polygon, may be only partial assessment
End Product as Raster, whole area.
1B
Table 1 is the comparison conventional and GIS

prospecting approach for the data driven case. The main
objective is same, to predict where the area should be had
a potential mineral deposit based on the known deposit
use the existing exploration data. Conventional
prospecting will consuming more time and effort with less
objectivity, may nor repeatable for the similar cases. In
another hand GIS prospecting method should be
repeatable doe to the criteria and algorithm are storage on
the computer memory. All area also will evaluate, not as
the conventional only as partial area which is able catch
by the visual interest (Setyadi, 2016).
Weight of evidence (WofE) is the machine-learning
data driven techniques to measure probability of data and
to estimate the posterior probability of the targeted
deposit. WofE is used to indicate the degree of correlation
between known mineral deposits (data training) and the
related factors of each fuzzy membership (BonhamCarter, 1996; Carranza, 2002; Kwang 2014; Shahi, 2014;
Keyan, 2015). The bayesian rule is expressed as:
{ | } =
SUM,
meaning in term of geology. Map should be analysis

(integrated) to use for mineral potential prediction. Map
analysis tool is use the mathematical method to combine
the map. Fuzzy gamma operator was selected for map
analysis. Fuzzy gamma operator is map combination
based on the weight of evidence (Bonham-Carter, 1997).
Regarding Lindsay (2014) and Yousefi (2014), fuzzy
gamma is the common operator to be used for mineral
prosperity modelling study, which is as a combination of
fuzzy algebraic sum and fuzzy algebraic product. Fuzzy
gamma operator for calculate the prosperity value could
be expressed as follow.
= 1 1 + 2 2 + 3 3 +
4 4 . . . . . . . . . . . . (2)
where:
P:
posterior
probability of each fuzzy
membership, : weight of map
P and come from fuzzy
model table
(}
{}
where P{BD} is the number intersected of fuzzy

member population as binary map {B} with the ore
deposit {D}.
Encoding or fuzzyfication is the process to convert
all the data input to the fuzzy membership from 0.1 1.0
based on the 10th percentiles data interval. This process is
to simplify the data become same level for all data set.
Weight of evidence (WofE) is the main machine to
evaluate what the data training model combination and
use to decode the data become binary map. The binary
map is the favourable mineral map, but they have no
Fig. 3: Mineral prosperity modelling algorithm
CASE STUDY
The study was taken over the Seruyung gold mining,
North Kalimantan, Indonesia.
Seruyung is a high
sulphidation ephithermal gold (HSE-Au) deposit, operated
by J Resources Nusantara. Mostly the area was covered by
62
PROCEEDINGS
the primary tropical forest. Strongly weathered was

limited to access geological outcrop, so the surface
geological map is not ideal for data modelling input.
Geology of Seruyung deposit is comprised of porphyritic
andesite unconformable covered by pyroclastic tuff.
Seruyung High Sulphidation Epithermal (HSE) gold
system is characterized by vuggy silica altered rocks
cantered on structurally controlled sulphide-rich
hydrothermal breccia lenses and surrounded by alunite
argillic and advance argillic alteration (Bautista, 2007;
Angeles, 2012). Lithology of hydrothermal breccia
associates with vuggy silica is the best geological
indicator for gold mineralization (Angeles, 2012). Good
alteration zoning and the availability of ground
geophysical survey data are very beneficial for the study
area.
Only surface geophysical data was considered has a
good quality to be used for data modelling. The
geophysical data consist of ground magnetic, IP and
resistivity survey. In general gold deposit was occurs on
the silica alteration zone which is coincide with high
resistivity anomaly. Destruction magnetic alteration
related to the hydrothermal alteration zone was mapped
with the low reduce to the pole (RTP) magnetic anomaly.
High IP-chargeability appears on the south-western part
was identify with the weak weathered diatreme breccia
with disseminated pyrite.
Figure 4 is present the conversion from raw RTP data
to the fuzzy-RTP in the MS Excel. Anomaly pattern is
relatively similar however the value is different. Red
colour as the high anomaly and blue colour as low
anomaly. Data conversion is based on the simple
classification rule using fuzzy table. The classification is
based on the statistical calculation using 10th percentile
formula in MS Excel.
The fuzzyfied data is the simpler data consist of ten
class fuzzy membership, so easier to be processed. WofE
is the probabilistic rule to create a fuzzy model of the
geophysical (input) data combination.
Figure 5 is
illustrate the RTP binary map (Figure 5.A) converted from
the fuzzy RTP (Figure 4.B) based on the fuzzy model
table (Figure 5A). Binary map is present the favourable
fuzzy membership correlate with the data training. All the
binary map (RTP, ANS, IP and RES) map produced from
the weighting process has no geological meaning yet.
Fuzzy gamma operator was selected to use to create the
prosperity map.
Every different mineralization zone represent by the
data training has unique model and has specific lithophysics distribution. All the different mineralization zone
should be mapped and modelled. Based on this study
identified nine different mineralization with different
fuzzy model. The prosperity modelling was create based
on those mineralization zone model.
Figure 4: Map conversion from raw RTP data to the fuzzyRTP
Using two different approach of binary map

integration returned two different map. Figure 5 (left) is
the alteration zone map. At least two kind of
mineralization style, porphyry and HSE with a total 9
mineralization zone and one intrusion. The mineralization
zone can use to help in temperature gradational
interpretation. Fuzzy algebraic SUM was used to integrate
all binary mineralization zone and resulted the mineral
prosperity map (Figure 5-right). This map present all the
potential mineral deposit. Red-purple (hot colour) indicate
the higher potential and green colour indicate the lower
potential. In comparison with the exploration drilling
result, seen all the high potential is correlates with the
high-longer gold intercepted from the exploration drilling.
The patchy low-moderate potential is occurs on the
peripheral zone related to the smaller ore bodies.
63
PROCEEDINGS
CONCLUSION
Model Fuzzy 1. MSZ

Peta RTP ANS IP RES
Gamma 0.45 0.48 0.66 1
Fuzzy
Posterior Probability
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0
0 1.00
0
0
0
0
0
0 0.66 0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.28 0.37 0
0
0.45 0.48 0
0
Statistical, such as fuzzy logic and weight of evidence

should be employed as an alternative tool to evaluate
exploration data thoroughly.
This method is able to uses for the detail mapping and
distinguished
physical properties of every
mineralization zone based on the available
geophysical data.
Based on this modelling study, suggested that the
prediction model is very close to the exploration
drilling result. The high prospectively is coincide with
the high grade drilling interception.
Understanding of different physical properties of
different alteration types is the important key to create
rule and criteria for geophysical alteration mapping.
Based on this study, it is predicted that Seruyung has
clearly mineralization zoning, increasing magnetic
response in the outer part due the reducing impact /
distance from the alteration source.
REFERENCES
Abedi, M., et al, 2013, Application of fuzzy AHP method to

integrate geophysical data in a prospect scale, a case study:
Seridune copper deposit, Bollettino di Geofisica Teorica ed
Applicata, Vol. 54, n. 2, pp. 145-164.
Abedi, M., et al, 2015, Mineral potential mapping in Central Iran
using fuzzy ordered weighted averaging method,
Geophysical Prospecting, Vol 63, pp. 461477
Angeles, C.A., (Jun), 2012, Evaluation Report of J resources
Ltds Seruyung Project, East Kalimantan, Indonesia, Interim
Report
Anshari, A.H., and Alamdar, K, 2009, Reduction to the Pole of
Magnetic Anomalies Using Analytic Signal, World Applied
Science Journal 7 (4), pp. 405-409
Asadi, H.H., et al., 2015, Exploration feature selection applied to
hybrid data integration modelling: Targeting copper-gold
potential in central Iran, Ore Geol. Rev
Bautista, B and Munajat, I., 2007, Seruyung Prospect - Project
Evaluation Summary, PT Sago Prima Pratama Interim
Report.
Barnett, C.T., William, P.M., 2006, Mineral Exploration Using
Modern Data Mining Techniques, Society of Economic
Geologist, Special Publication 12, pp 295-310.
Clark, D.A., and Emerson, D.W., 1991, Notes on Rock
Magnetization Characteristics in Applied Geophysics
Studies, Exploration Geophysics vol 22, pp. 547-555.
Hoschke, T., 2015, The Geophysics of Cu-Au Porphyry System
and Epithermal Au Deposits, MGEI Presentation
MacLeod, I.N, Jones, K and Dai, T.F.,1993, 3-D analytical
Signal in the Interpretation of the Total Magnetic Field Data
at Low Magnetic Latitude, Exploration Geophysics, vol 24,
pp. 679-688
Moon, C.J., et all, 2006, Introduction to Mineral Exploration,
Second Edition, Black Well Publishing
Keating, R and Sailhac, P, 2004, Use of the analytic Signal to
Identify Magnetic Anomalies due the Kimberlite Pipes,
Geophysics Vol.69 No 1, pp.180-190
Keyan, X,et.all, 2015, GIS-based 3D Prospectivity Mapping: A
Case Study of Jiama Copper-Polymetallic Deposit in Tibet,
China, Ore Geology Reviews.
Payne, C.E., et al., 2014, From 2D to 3D: Prospectivity
Modelling in the Taupo Volcanic Zone, New Zealand, Ore
Geol Rev.
Rajagopalan, S, 2003, Analytic signal vs. reduction to pole:
solutions for low magnetic latitudes, Exploration
Geophysics 34(4), pp.257 262
Figure 5: Binary RTP map (Bottom) created from the fuzzy

model (Top)
Figure 5: The output from the mineral predictive modelling.

Top: is the alteration zone modelling and bottom is the mineral
prosperity map
64
PROCEEDINGS
Setyadi, H, 2016, Model Prospeksi Endapan Mineral Berbasis

Domain Geologi, Studi Kasus Endapan Emas Tipe HSE-Au,
Seruyung, Institut Teknologi Bandung (on preparation)
Shahi, A and Kamkar-Rouhani, A., 2014, A GIS-based weightsof evidence model for mineral potential mapping of
hydrothermal gold deposit in Torbat-eHeydarieh area,
Journal of Mining & Environment Vol 5, No.2, pp 79-89.
William-Jones and Clark, J.R, 2012, Transport and Deposition of
Gold in Magmatic Hydrothermal Systems, Department of
Earth and Planetary Sciences McGill University, Montreal,
Canada
Zonge, K.L. and Hughes, L.J.,1980, The Complex Resistivity
Method, Zonge Engineering & Research Organization, Inc,
Tucson USA.
ACKNOWLEDGEMENT
I would like to say thank to the J. Resources
Nusantara management, PT Sago Prima Pratama
geologists who have given the opportunities to review and
evaluate as well as to publish their data for this study case.
Appreciate MGEI committee to the opportunity for
present and publish this paper. Appreciation are also to the
anonymous reviewers for their constructive comments.
65
PROCEEDINGS
Geobotanical Remote Sensing for Mineral Exploration in Thick Vegetation Areas

N. H. Hede1*, K. Koike2, K. Kashiwaya2, S. Sakurai3, R. Yamada4, and D. A. Singer5
1
Earth Resources Exploration Research Group, Faculty of Mining and Petroleum Engineering, Institut Teknologi
Bandung, Jl. Ganesha 10, Bandung 40132, Indonesia. Corresponding author: naftali@mining.itb.ac.id
2
Department of Urban Management, Graduate School of Engineering, Kyoto University, Katsura C1-2-215, Kyoto 6158540, Japan
3
Graduate School of Advanced Integrated Studies in Human Survivability, Kyoto University, Kyoto 606-8306, Japan
4
Graduate School of Environmental Studies, Tohoku University, Sendai, Miyagi 980-8578, Japan
10191 N.
5
Blaney Ave., Cupertino, CA 95014, United States
ABSTRACT
Remote sensing optical sensors onboard satellites have been effectively used in geological fields to identify minerals and
rocks via reflectance and emissivity spectral characteristics of earth surface materials. However, as for the spectral
applications, traditional remote sensing for mineral exploration researches has been mostly limited to arid and semi-arid
areas where vegetation is sparse or absent. Vegetation is the most critical barrier for geologic identification and mapping
because reflectance spectra of vegetation conceal the spectra of underlying soils and rocks. One key factor of geobotanical
remote sensing for mineral exploration is the presence of unusual vegetation in mineralized areas, which can be identified
from the reflectance spectral pattern of plants. Based on this background, this paper discusses a vegetation index (VI) for
detecting vegetation anomalies using reflectance data at several bands in the visible to near infrared and shortwave
infrared regions so that the VI was sensitive to vegetation stress which may be caused by metal absorption. At first, a set of
laboratory experiments was undertaken to clarify the relationship between metal contents in soils and reflectance spectra of
a selected plant species. The new VI was applied to Landsat ETM+ images of two mineralized areas containing
hydrothermal copper deposits and kuroko deposits. Through these case studies, a new level of understanding of
geobotanical remote sensing and the method that was successfully applied for detecting latent ore deposits in a wide, dense
forest area. is discussed.
1987). Metal-induced vegetation stress interferes with
chlorophyll activity and inhibits water absorption from
soils and the supply to leaves (Barcel & Poschenrieder,
1990). Furthermore, an understanding of the relationship
of plant formations to geological conditions and metal
content in soils is essential. In the optical region, the
wavelength ranges from visible to near infrared (VNIR),
4001,400 nm, in which the effect of absorbed metal on
the reflectance spectra is easy to identify; this approach
has been used in GBRS. This effect has also been
observed at longer wavelengths, in shortwave infrared
(SWIR) bands of 1,4002,500 nm as the vegetation stress,
and it is likely to be enhanced by combining VNIR and
SWIR (Horler et al., 1980).
Vegetation index (VI) is considered an important
approach to analyzing vegetation stress using remote
sensing (Slonecker, 2011), which is a mathematical
manipulation of a digital number used to quantify and
detect vegetation conditions. The VI is used to emphasize
the change in reflectances at bands selected for estimating
the magnitude of stress. Common to numerous proposed
VIs, most use VNIR data only. However, SWIR
reflectances are sensitive to leaf water content and thus
can be used to detect vegetation stress by water supply
interference. Therefore, a VI derived from reflectance data
in the VNIR and SWIR regions is expected to enhance
detection accuracy of vegetation anomalies. Based on this
expectation, the Vegetation Index considering Greenness
and Shortwave infrared (VIGS) is proposed (Hede et al.,
2015). This index is aimed at wide availability to general
INTRODUCTION
The consumption of mineral resources has
experienced rapid growth in many countries in recent
years which has led to a higher demand and resulted in
considerable needs for innovation in natural resource
exploration. Remote sensing is one of the most popular
tools for indirect exploration and is commonly used
during the prospecting phase as it can provide a rapid
assessment at a low cost and with minimal risk. As for
spectral applications, the validity of remote sensing is
limited to arid and semiarid areas where vegetation is
sparse or absent. Because the reflectance spectra of
vegetation conceal the spectra of underlying soils and
rocks, vegetation is the most critical barrier to geological
identification and mapping in the application of remote
sensing.
However, vegetation can be an essential part of
geological researches. The scientific study of the
relationship between vegetation and geological condition
is called geobotany. Traditional geobotany is an extension
of geochemistry and biogeochemistry, resulting in a field
able to fully examine botanical phenomena related to
geological influence (Sabins, 1999). In recent years, the
integration of geobotany with remote sensing technology,
geobotanical remote sensing (GBRS), has allowed for the
study of spectral response patterns relating to
morphological and physiological changes resulting from
the absorption of metals (the term metal in this study
refers to heavy metals) in vegetation (Bruce & Hornsby,
66
PROCEEDINGS
multispectral satellite imagery by integrating visible

green, red, NIR, and SWIR reflectances as:
GR
N R
N S1
N S 2 (1)
VIGS = w
+w
+w
+w
G + R
N +R
N + S1
reflectance spectra to VNIR and SWIR band reflectances

of Landsat TM and ETM+ sensors, original reflectances
were averaged over the wavelength range of each band.
N + S2
LABORATORY EXPERIMENT RESULTS
where G, S1, and S2 denote reflectances in the

visible green and two SWIR regions, respectively, and w1,
w2, w3, and w4 are weights for emphasizing each term. For
Landsat ETM+ imagery, S1 and S2 correspond to bands 5
and 7 (B5 and B7). Considering reports on the variability
of reflectances induced by vegetation stress in the selected
regions, a weights set, w1=1.0, w2=0.5, w3=1.5 and w4=1.5,
was revealed as the most suitable because the VIGS from
this combination greatly enhanced the difference in stress.
The effect of metal soil contamination appeared in

the total leaf chlorophyll content, which generally
decreased with metal content. The effect of metal content
on total chlorophyll content was strongest in the Cd
specimens. The Cu and Zn specimens had similar values
and trends of total chlorophyll content with the variation
of content and days. On the contrary, this effect was weak
in the Pb specimens, in which there was no significant
difference of total chlorophyll content despite change in
Pb content. Using the Landsat band reflectances, the
VIGS is calculated and their changes in metal content and
elapsed day are compared in Fig. 1. A change in VIGS
values appeared in the Pb specimens, and more
remarkable changes were revealed by the VIGS in the Cd
and Zn specimens. Another noteworthy point is that the
magnitude of VIGS values accurately followed the order
of metal content for most elapsed days; i.e., the VIGS
values increased in order of low, medium, and high
content. This consistency verifies the effectiveness of
using SWIR for the VI.
DATA AND METHOD

This study consists of laboratory experiments and
case studies. The standardized experiments involved
cultivating Japanese mustard spinach (Brassica rapa var.
perviridis) to clarify the relationship between vegetation
response and metal contamination by analyzing the
response of vegetation grown in soils contaminated by
selected metals (Hede et al., 2015). This part is
indispensable to identifying vegetation anomalies caused
by metal contamination. The case studies were conducted
in the Jambi region and Hokuroku region. The Jambi
region features copper mineralization and the Hokuroku
region in Japan is one of the worlds most famous sites of
Volcanogenic Massive Sulfide (VMS) ore deposits of the
kuroko (black ore) type. Both locations were selected
based on the availability of geo-exploration datasets,
including geochemical data and mining activity.
The Japanese mustard spinach (Brassica rapa var.
perviridis) was selected for the experiment because of its
relatively short lifetime (approximately 24 months) and
ease of cultivation from seed under laboratory conditions.
Four types of metal, Cu, Pb, Zn, and Cd, were selected for
simulating contaminated soils for the following reasons.
Cu, Pb, Zn are common metals present in mineralized
soils around metal deposits. Although soil Cd, a
contaminant derived from mineral outcrops, is rare and
generally limited when associated with sphalerite (ZnS) in
the form of greenockite (CdS), it has been recognized as
having strong biological toxicity. In total, 26 pots were
prepared: 2 pots for each of the four metals and the three
contaminant levels (24 pots) and, as a control, 2 pots of
normal soil without metals added. The seeds were
germinated in the control soil for 2 weeks, and then the
sprouts were transplanted to the pots with metalcontaminated soils. The plants had been exposed to light
and dark for 16 and 8 hours a day, respectively and
supplied with distilled water daily to preserve the moisture
level. The measurements of reflectance spectra and total
chlorophyll content began when the plants bore four or
more leaves at 60 days after the sowing and continued
over the next 20 days, at 5-day intervals. To convert
Fig. 1 Comparison of NDVI and VIGS change with metal

contents and days since sowing. Each VI value is an average of
two specimens at each metal content in soils and two reference
specimens without metal addition (Hede et al., 2015).
APPLICATION TO HYDROTHERMAL COPPER DEPOSIT

AREA
The first study area in central Sumatra (western
Indonesia) was of 5050 km size (Fig. 2), mostly
mountainous with thick vegetation cover. The study area
is mainly overlain by block-shaped Permian intrusions.
This block is the most important to mineral exploration
because many mineralized regions have been found in this
formation. The subordinate main geology is Triassic
Cretaceous granitoid. The study area is characterized by
joint or separate occurrences of precious and base metals
in places (Crow et al., 1993).
Content data of Cu, Pb, and Zn in near-surface rocks
and soils at 282 sampling points (British Geological
Survey, 2007) were transformed into logarithms to reduce
bias of content distribution for subsequent geostatistical
analyses.
After
approximating
experimental
semivariogram by a model, simple kriging (SK) and
ordinary kriging (OK) were used for estimating metal
contents across the study area depending on higher
estimation accuracy. As the result, distributions of Cu and
67
PROCEEDINGS
Pb contents were estimated by OK and Zn by SK (Fig. 2).

These metal contents are found to be strongly correlated
with the geology and controlled by the faults. The high Cu
and Pb contents showed clear continuity along the strike
of major NWSE faults. This trend was weak for the Zn
contents, which were high along the NS direction.
Fig.2 Distributions of metal contents estimated by ordinary

kriging (OK) for Cu and Pb and simple kriging (SK) for Zn.
Black lines and circles here show faults and metalliferous
mineral occurrences, respectively (Hede et al., 2015).
A subscene of a Landsat ETM+ image on 15 Aug.

2002 was selected for the VI analysis. Reflectance of each
pixel was derived from the original image by an
atmospheric correction and a topographic correction.
VIGS was calculated using the corrected reflectance and
distributions of VIGS values were overlain on the
geologic boundaries and faults (Fig. 3). By defining a
value greater than mean (m) + standard deviation () as
VIGS anomaly, the concentration-area (C-A) model was
applied to the density of anomalies, to clearly distinguish
large-value zones (Fig. 4a) in which the anomaly densities
were categorized as I: low, II: medium, and III: high.
High-density areas are mainly covered by the primary
forest. Furthermore, the spatial distribution of Cu contents
classified as low, medium, and high was correlated with
the density of VIGS anomalies (Fig. 4a). Obviously, the
high Cu-content data are in the high-density category in
the northern half of the study area (marked by A in Fig.
4a). This relationship did not appear in the Ngaol region,
because that region is covered by the secondary forest and
paddy fields which generally have small VIGS values.
Fig.3: Distribution of VIGS from Landsat ETM+ image,

superimposed on geologic boundaries and faults (Hede et al.,
2015).
High densities are found in the Tertiary intrusive

rocks as shown by the B in Fig. 4a. This is a reasonable
tendency, because metal contents are generally higher in
intrusive or volcanic units. The C-A model was also used
to correlate the values with the metal content data in
detail. As a result, average metal contents increase with
68
PROCEEDINGS
category, for all three metals (Fig. 4b); large VIGS values
are associated with high metal contents, and small values
with low contents. This trend confirms the satisfactory
VIGS correspondence to metal soil contamination and the
applicability of VIGS to mineral exploration in areas with
thick vegetation using the CA model.
Figure 4: (a) Distribution of (I) low, (II) medium, and (III) high
density categories of VIGS anomaly values from the C-A model,
overlaid by). The content classifications follow the C-A model in
Fig. 5. (b) Relationships of Cu, Pb, and Zn contents with VIGS
categories, shown by box plots (Hede et al., 2015).
APPLICATION TO A KUROKO DEPOSITS AREA

The Hokuroku district extends over 4040 km2 in
Akita Prefecture, northern Japan and is known to be the
most thoroughly studied kuroko deposit area in the world.
Kuroko is one type of volcanogenic massive sulfide
(VMS). Most of the study area (Fig. 5a) is underlain by
middle Neogene formations consisting of volcanic and
sedimentary rocks. There was active volcanic activity
during the Neogene time, termed the Green-Tuff
movement. Kuroko deposits were formed during the
Miocene, Nishikurosawa age and hosted in the volcanic
rocks composed of dacitic to rhyolitic lavas, lithic tuff
breccias, and tuffs. The Hokuroku district is located in a
subsided sedimentary basin formed by acid volcanic
activity during the Miocene time in the Green-Tuff
movement. Therefore, genesis of these kuroko deposits is
considered to be associated with fracturing in the
basement composed of pre-Tertiary phyllite, slate,
sandstone, and chert, and active eruptions of basaltic and
andesitic rocks along these fractures (Yamada & Yoshida,
2013).
Fig.5 Top (a) Locations of mines in the Hokuroku district

overlaid with a true color composite image of Landsat ETM+
acquired on 25 Jul. 2002. Size of the circle is different with the
deposit magnitude. Bottom (b) Spatial characterization of
standardized VIGS anomaly (greater than mean + standard
deviation) density to enhance the anomalous zones that can be
correlated with the mines.
Five scenes of Landsat ETM+ images acquired on 5

Jul. 2002, 20 Aug. 2000, 5 Sep. 2000, 21 Sep. 2000, and
69
PROCEEDINGS
13 Oct. 2002 were used to calculate the VIGS values

following the procedures as above mentioned. In the
kuroko deposits area, temporal change in the VIGS values
can be used as a key indicator. Therefore, m of VIGS
values of the five scenes divided by was used to detect
VIGS anomalies. The CA model was applied to the (m
) anomalies, and the high density zones are highlighted
by reddish colors as shown in Fig. 5b. It is noteworthy that
the high densities zones draw ring-shaped structures on
which the mines are located. The ring structure suggests
submarine resurgent cauldron that had genetic relation
with the kuroko deposits (Kouda & Koide, 1978). Several
mines such as Matsumine do not correspond to the high
density because of little vegetation around them.
Consequently, VIGS is also applicable to VMS-type areas
with thick vegetation.
correlated with the geology and controlled by the faults

through a kriging interpolation of the point content data.
The effectiveness of VIGS was proved by this case study
because the VIGS anomalies were appeared on the high
content zones common to the three metals. This
concordance is probably originated from that the plant
formations (mainly the primary forest) in the high metal
zones are closely related to the geological units. VIGS
anomalies were also appeared around the developed mines
of kuroko deposits in the Hokuroku district. A noteworthy
feature was that the anomalies form ring-shaped structures
on which the mines were located. Those results verify the
applicability of VIGS to metal exploration under the
vegetation.
REFERENCES
Barcel, J., & Poschenrieder, C., 1990, Journal of Plant
Nutrition.
British Geological Survey, 2007, Keyworth, Nottingham, UK.
Bruce, B., & Hornsby, J. K., 1987, Geocarto International.
Crow, M. J., Johnson, C. C., McCourt, W. J., & Harmanto, 1993,
Special Publication of the Directorate of Mineral Resources,
No. 52-B, Bandung.
Hede, A. N. H., Kashiwaya, K., Koike, K., & Sakurai, S., 2015,
Remote Sensing of Environment, 171, 8397.
Horler, D. N., Barber, J., & Barringer, A. R., 1980, International
Journal of Remote Sensing, 1(2), 121136.
Kouda, R. & Koide, H., 1978, Mining Geology, 28, 233244.
Sabins, F. F., 1999, Ore Geology Reviews, 14(3-4), 157183.
Slonecker, E. T., 2011, In P. S. Thenkabail, J. G. Lyon, & A.
Huete (Eds.), Hyperspectral remote sensing of vegetation,
CRC Press, 561578.
Yamada, R. & Yoshida, T. (2013) Journal of the Geological
Society of Japan, 119, Supplement, 168179.
CONCLUSIONS
A new vegetation index VIGS, combining
reflectances in the VNIR and SWIR regions, was
proposed for detecting vegetation stress caused by metal
soil contamination in densely vegetated and mineralized
areas. A set of laboratory experiment using Japanese
mustard spinach is undertaken to investigate the change in
reflectance spectra depending on the concentration of
selected four metals (Cu, Pb, Zn, and Cd) in soils. The
results demonstrate that the VIGS values can enhance the
stress difference depending on the metal contents in soils.
VIGS was furthermore examined for two mineralized
areas containing hydrothermal copper deposits in Jambi,
central Sumatra, Indonesia and kuroko deposits in the
Hokuroku district, northern Japan. In the Jambi area, the
Cu, Pb, and Zn contents were found to be strongly
70
PROCEEDINGS
71
PROCEEDINGS
An update of key characteristics of Awak Mas mesothermal gold deposit, Sulawesi

Island, Indonesia
Ernowo1,2, Franz Michael Meyer2, Arifudin Idrus3, Heru Widyanarko4 and Nita Lely Endrasari4
1
Geological Agency of Indonesia, Bandung, Indonesia

Institue of Mineralogy and Economic Geology (IML), RWTH Aachen University, Germany.
3
Gadjah Mada University, Yogyakarta, Indonesia
4
PT. Masmindo Dwi Area, Jakarta, Indonesia
ABSTRACT
Awak Mas is located within the metamorphic belt of Sulawesi island forming part of the Creataceous Latimojong formation
consists of flysch sequence that is locally intruded by diorite dykes, transected by NNE-SSW trending, parallel to subparallel and sub-vertical fault zones.
At the cut-off grade of 0.5 g/t Au, Awak Mas was estimated to contain a measured and indicated resources of 45.1 Mt at
1.30 g/t Au and additional inferred resources of 6.5 Mt at 1.13g/t Au for a total resource of 2.13 Mi Oz gold with proven
and probable reserve of 20.2 Mt at 1.58 g/t Au equals to 1.03 Moz
The Awak Mas gold deposit is hosted by phyllite and schist that represent the metamorphic products of shale, Fe-rich shale
and wacke protoliths. Metamorphic chlorite thermometry indicates that metamorphism conditions reached up to 300oC,
which correspond to lowgrade greenschist facies.
Mineralized quartz-ankerite-albite veins are enveloped by zones of hydrothermal alteration with a proximal albite-ankeritepyrite assemblage and distal zone comprising muscovite-albite-chlorite. Muscovite occured finely distributed as sericite
while quartz, albite, and ankerite form continuous and discontinuous veins and/or veinlets.
Two main styles of mineralization include quartz veins and hydraulic breccias with sulfides and a sodium-rich fluid
alteration assemblage. Euhedral pyrite is the most abundant sulfide and disseminated in the albite-quartz-ankerite
alteration zone with variable grain sizes of up to 1 mm. Galena and chalcopyrite commonly occurred as inclusion in pyrite.
Assay data of drill core samples reveal gold grades in the mineralized phyllites and schists ranging from 0.02 to 0.54 g/t
and from 0.2 to 3.8 g/t, respectively. Elevated Au values are generally found in the hydraulic brecciated schist. Micron-size
gold grains were detected as inclusion in pyrite and interstitial between pyrite grain boundaries have the gold fineness of
92.56 which implies the characteristics of common orogenic gold deposit.
Fluid inclusion phases are dominated by aqueous fluids with additional scarce monophase CO2 inclusions.
Microthermometry analysis of primary aqueous fluid inclusions gives an estimated salinity of 2 to 5 wt.% NaCl eq. and
homogenization temperatures of 200 to 300oC. The oxygen isotope composition of the ore fluid yielded 18O values ranging
between 9 and 11.2 which is not indicative of a definite fluid source. The fluid halogen Br/Cl and I/Cl ratios appear to confirm the
presence of a metamorphic fluid derived from original marine sediments. Thus, the characteristics of the Awak Mas gold deposit
are consistent with those of the mesothermal orogenic gold concept.
Keywords: Awak Mas, metamorphic fluid, mesothermal,
orogenic gold
flysch sequence that is locally intruded by diorite dykes.
Tectonically, the area is transected by NNE-SSW
trending, parallel to sub-parallel and sub-vertical fault
zones (Querubin&Walters,2012).
The Awak Mas hosts the Salu Bulo, Tara and Awak
Mas gold prospects. At the cut-off grade of 0.5 g/t Au,
Awak Mas gold prospect was estimated to contain a
measured and indicated resources of 45.1 Mt at 1.30 g/t
Au and additional inferred resources of 6.5 Mt at 1.13g/t
Au for a total resource of 2.13 Mi Oz gold with provent
and probable reserve of 20.2 Mt at 1.58 g/t Au equals to
1.03 Moz
INTRODUCTION
For many decades, exploration and extraction of gold
in Indonesia is focused on the volcanic rock-hosted
deposits. Based on Indonesian Metallogenic Map
(Harahap 2014), from approximately 194 well known
gold deposit types in Indonesia, more than 90 % are of
epithermal gold deposits hosted by volcanic rock, 6.2 %
porphyry and skarn, the rest are VHMS and Carlin types.
Present-day, however, gold exploration activities are more
emphasized on the metamorphic rock-hosted deposits
particularly in the eastern part of Indonesia. One of the
metamorphic rock-hosted gold deposits discovered in
Indonesia is Awak Mas prospect.
Awak Mas is located within the metamorphic belt of
Sulawesi island forming part of the Creataceous
Latimojong formation. This is made up by phyllites,
slates, mafic to intermediate volcanics, limestones, and
schists representing a platform and/or fore arc trough,
DATA AND METHOD

This study was carried out by analysis of secondary
and primary data. Secondary data were taken from
published papers and unpublished technical reports from
the PT. Masmindo Dwi Area. Primary data were collected
by fieldwork, sampling and laboratory analyses.
72
PROCEEDINGS
ankerite-pyrite, characterized by progressive replacement

of the muscovite by albite (Fig.2).
The hydrothermal alteration process forming
albitization can be formulated by the mineral equilibria
reaction of muscovite plus sodium-rich hydrothermal
fluids to form the albite below:
The petrographic observation and preliminary

mineral identification were done on polished thin sections
using a Zeiss microscope. Mineral chemistry of the main
components from host rocks and ore minerals were
determined using a JEOL-JXA-8900R electron probe
micro-analyzer (EPMA). Mineral species were identified
by QEMSCAN. Microthermometry of fluid inclusions
was analyzed by LINKAM THMS600 heating and
freezing stage. All of these measurements were conducted
at the Institute of Mineralogy and Economic Geology,
RWTH Aachen University, Germany. Crush-leach
analysis of fluid inclusion was done at Leoben University,
Austria. Rock geochemistry was analyzed at ActLab,
Canada.
Samples of vein quartz and pyrite were roughly
crushed, then hand-picked for purity under technical-grade
ethanol using a binocular microscope for oxygen and
sulphur
isotope analyses at Tbingen University,
Germany.
KAl2(AlSi3O10)(OH)2 + 3Na+ + 6H4siO4 3NaAlSi3O8 +

K+ +2H+ +H2O.
Muscovite is finely distributed as sericite while

quartz, albite and ankerite form continuous and
discontinuous veins and/or veinlets. Banded graphite
occured as accessory mineral in the proximal zone.

Host Rock Lithology
The Awak Mas gold prospect is hosted by red
hematitic phyllite, green chloritic phyllite, dark phyllite
and mica schist that represent the metamorphic products
of shale, Fe-rich shale and wacke protoliths (Fig. 1). Host
rocks are predominantly made up of variable grain sizes of
quartz, muscovite, albite, quartz, and chlorite showing
lepidoblastic, granoblastic and porphyroblastic textures.
Metamorphic chlorite thermometry indicates that
metamorphism conditions reached up to 300oC, which
corresponds to lowgrade greenschist facies.
Fig. 2. Photomicrograph showing the relict of muscovite

replaced by albite
Gold Mineralization
Two main styles of mineralization include quartz
veins concordant and discordant to the foliation and
hydraulic breccias with sulfides and a sodium-rich fluid
alteration assemblage. Euhedral pyrite is the most
abundant sulfide and occurs disseminated in the albiteankerite-pyrite alteration zone with variable grain sizes of
up to 1 mm. Galena and chalcopyrite are commonly in
form of inclusions in pyrite. Micron-size gold grains were
detected as inclusion in pyrite and interstitial between
pyrite grain boundaries (Fig. 3).
Assay data of drill core samples reveal gold grades in
the mineralized phyllites and schists ranging from 0.02 to
0.54 g/t and from 0.2 to 3.8 g/t, respectively. Elevated Au
values are generally found in the hydraulic brecciated
schist. There is no clear relationship between gold and
base metal contents. However, samples with high gold
values tend to contain low copper, lead and zinc.
The elemental composition of 90 m long sized
fracture filling gold grain has 92.35% Au and 7% Ag with
Au-Ag ratio 13.2 and gold fineness of 929.56. The gold
fineness is an indicative of common orogenic gold
(Grooves et al,2003)
Fig. 1. Chemical classification for terrigenous clastic sediments

(after Herron, 1988)
Hydrothermal Alteration
The mineral quantification from selected phyllite
representing the alteration zone from the least altered rock
in distal to proximal zone shows the mineral assemblages
of muscovite-albite-chlorite-quartz of least altered distal
rock in distal zone. Albite-chlorite-calcite typifies the
middle zone and in the proximal comprising albite-
Ore Fluid Characteristics

Fluid inclusion phases are dominated by aqueous
fluids with additional scarce monophase CO2 inclusions.
Heating-cooling experiments were performed on primary
73
PROCEEDINGS
aqueous fluid inclusions from discordant quartz-ankeritealbite veins yielded ice melting temperature of -0.7 to 3.7oC gives a calculated salinity of 2 to 5 wt.% NaCl eq.
The homogenization temperatures vary from 200 to
300oC.
Fig.5. Oxygen isotope values of important geological reservoirs

(After Hoefs, 1997)
Crushed leach analysis was conducted to quartz grains

from both of veins and breccia. The ratio of halogen
chlorine (Cl), bromine (Br) and iodine (I) provide
alternative fluid parameters. The Br/Cl ratio and I/Cl ratio
(Fig. 6,7) are closer to bulk earth than to seawater and
distinct from magma fluids indicate a dominant crustal
fluid source. The high I/Cl ratios indicate a fluid sourced
from sediments, possibly rich in organic material
(Goodwin, 2010). Thus, halogen ratios appear to confirm
the presence of a metamorphic fluid derived from original
marine sediments.
Fig. 3. Backscatered electron image showing a micron-size gold

grain associated with chalcopyrite and galena within pyrite
crystal.
The 34S values show a relatively large range from

0.6 to 12.86 CDT can not be allocated to a single
sulphur source (Fig.4). O-isotopic fluid composition
calculated for T of 250 oC (Zheng, 1993) resulted a
narrow range of 18 O values 9-9.3 SMOW for
concordant quartz vein, 9-9.9 SMOW for discordant
quartz vein and 9-11.2 SMOW for breccia. These 18O
composition of ore fluid is not indicative of a definite
fluid source (Fig.5).
Fig. 6. Crustal fluid halogen chemistry (after Wilkinson, 2001)
Fig. 4. Sulphur isotope composition of some deposit types

(after Wulff, 2008)
Fig. 7. Log-log plot of halogen ratios determined in quartz

samples (after Goodwin, 2010)
74
PROCEEDINGS
Grooves et al, 2003, Economic Geology 98, 1-29.

Harahap et al, 2014, Indonesian Metallogenic Map.
Herron, M.M., 1988, Journal of Sedimentary Petrology 58, 820829.
Querubin &Walter, 2012, Majalah Geology Indonesia 27, 69-85.
Wilkinson,J.J., 2001, Lithos 55, 229-272.
Wulff, K., 2003, PhD Thesis, RWTH Aachen University.
Zheng,Y.F., 1993, Geochimica et Cosmochimica 57, 1079-1091.
CONCLUSIONS
Gold mineralization of Awak Mas prospect is hosted
by sequences of sedimentary rocks subjected to a low
grade metamorphism as phyllite and schist, associated
with pervassive hydrothermal alteration overprint resulted
in proximal alteration zone with albite-ankerite-pyrite
assemblage. The oxygen isotope and halogen composition
of fluid inclusions suggest that hydrothermal fluids were
predominantly sourced from metamorphic dewatering
reactions of marine sediments. The alteration
asssemblages, mineralogical composition of gold, and ore
fluid characteristics suggesting that the Awak Mas gold
deposit are consistent with those of the mesothermal
orogenic gold deposit.
ACKNOWLEDGEMENTS
This paper is written in frame of the first author PhD
work at RWTH Aachen University. PhD Scholarship is
provided by Ministry of Energy and Mineral Resources of
The Republic of Indonesia. Those are appreciated and
deeply thankful. A deep gratitude goes to the management
of PT. Masmindo Dwi Area, the owner Awak Mas
prospect for their permission to access study area and
data.
REFERENCES
Goodwin,N.R.J., 2010, PhD Thesis, University of Manchester.
75
PROCEEDINGS
Discovery, Geology and Origin of the Lakuwahi Volcanogenic Au-Ag-Pb-Zn

Deposit, Romang Island, eastern Indonesia.
Joseph Ogierman1
5)
Exploration Manager, PT Gemala Borneo Utama

ABSTRACT
Indonesia is well known for epithermal Au and porphyry Cu-Au deposits in island arc settings but less known for island arc
VMS mineralisation such as the Au-Ba-rich volcanogenic deposits on Wetar and Romang Islands in the Banda Arc. Au/Ag
mineralisation was discovered on south Romang by Billiton in the 1990s in a large magnetic anomaly, outlined by aerial
surveys and interpreted to reflect intense hydrothermal alteration. Mapping and soil sampling defined a Au-Ag-Ba anomaly
which was tested by 14 diamond drillholes. Although several significant intersections were returned they chose not to
proceed with the project. PT GBU were granted tenements on Romang in 2006 and since then has outlined a large
polymetallic mineral resource at Lakuwahi of 82Mt with 1Mozs of gold, 68 Mozs silver and just over 1 Blbs of both lead
and zinc (JORC reporting standard). In addition, the polymetallic deposit is partially overlain by a high-grade manganese
oxide deposit of 711 kt @ 43.5% Mn (JORC).
Lakuwahi consists of a cluster of mineral deposits hosted by andesitic Lakuwahi Volcanics formed in a caldera atop a
shallow submarine volcanic edifice. Each deposit is associated with one or more sub-vertical barite-rich breccia feeder
zones. Multiple hydrothermal events are evident beginning with widespread silicification. Subsequent brittle deformation
created zones of high permeability, allowing hydrothermal fluids to deposit Au + Ag + barite + base metal sulphides in
breccias and stockwork veins within feeder zones and along sub-horizontal zones, below the seafloor surface. Fluids also
vented onto the seafloor to form barite-rich exhalative layers, variably mineralised with Au, Ag and base metals. Lakuwahi
Volcanics and exhalites were subsequently covered by volcaniclastics and a reef limestone. The hydrothermal system then
re-activated, with fluids replacing limestone with Mn / Fe Oxides +/- As-Sb-Tl. The system is still weakly active with small
sulphur fumeroles venting at two locations.
Continued uplift in the past 1-2Ma has caused the volcanic edifice to became emergent to form Romang Island. Block
faulting resulted in some deposits becoming exposed by erosion while others were down-faulted and covered by lagoonal
sediments. Near-surface deposits have an Au-Ag rich oxide zone. Preliminary economic assessment indicates a viable
mining project with conventional CIP treatment of the oxide and transition zones over a 10 year mine life producing 20,000
ounces Au and 600,000 ounces Ag per annum. Metallurgical and process modelling is currently being finalized with a plan
to start mine construction in late 2016. Due to prevailing Indonesian mining regulations including restrictions on the export
of metal concentrates, it is currently uneconomic to develop the larger, underlining sulphide polymetallic deposits.
the whole of south Romang Island, only 6 rock chip
samples were collected, one of which returned 610 g/t Ag
but no further work was undertaken and the tenement was
relinquished in 1992.
Billiton, operating as PT Prima Wetar Mining, arrived
in 1997, looking to expand the resource base of their gold
mine on nearby Wetar Island. They flew an aerial
magnetic survey, outlining several magnetic lows,
interpreted to reflect hydrothermal activity. The largest of
these became the Lakuwahi Project (fig 1). Billiton had
discovered gold + barite-rich VMS-style mineralisation on
Wetar and were able to recognize barite-rich outcrops on
south Romang as potentially representing a similar
geological setting.
Billiton followed up the Lakuwahi Anomaly with
CSAMT resistivity surveys, detailed soil geochemical
surveys and finally with a scout drilling program of 14
drill holes. The program was largely successful with every
hole except one intersecting either >1 g/t Au or >100 g/t
Ag with the most significant intersection of 47m @ 1.45
g/t Au + 26 g/t Ag from surface. Even so, when mining
finished on Wetar, Billiton ceased all gold exploration in
the Banda Arc.
The prospect lay dormant until 2006 when Jakartabased PT Gemala Borneo Utama (GBU) were granted
IUP exploration tenements covering north and south
Romang. GBU entered a JV with ASX-listed Robust
Resources in early 2008 allowing exploration activities to
INTRODUCTION
Precious metal-bearing volcanogenic massive sulphide
(VMS) deposits are a sub-type of VMS deposit
(Hannington et al. 1999). Only one example of this style
of deposit in Indonesia, on Wetar Island, has been
described in the literature. A second significant deposit
can now be added, the Lakuwahi Deposit on nearby
Romang Island.
This paper describes exploration history including
techniques used and documents mineralization styles and
mineralogy of Lakuwahi through drill core analysis,
petrography and preliminary fluid inclusion and
geochronology work. Similarities and differences between
Romang and Wetar deposits are noted and genetic models
are proposed to explain the differences between the two.
EXPLORATION AND DISCOVERY
Discovery of the Lakuwahi deposit is a case of
exploration persistence. Anomalous Au/Ag samples were
first collected on Romang Island in the early 1990s but it
has taken 25 years for a mining project to reach feasibility
study stage.
The first company to undertake modern exploration, PT
Nailaka Marhila Mining, focused on the north half of
Romang in 1991. Using reconnaissance geochemical
sampling they discovered a narrow Au-Ag-Pb-Cu
epithermal vein, tested by trenching and 18 drillholes. For
76
PROCEEDINGS
start on the island with drilling at Lakuwahi

recommencing in late 2008.
Billitons work showed that Ag in soil is an important
vector for outlining polymetallic mineralisation. Based on
this, GBU infilled and extended the Billiton work on a 60
x 30m grid, collecting over 6000 samples. The soil
program confirmed a close correlation between
polymetallic mineralisation and anomalous Ag with a
>3ppm Ag contour closely outlining the major Lakuwahi
deposits.
Unfortunately Billitons CSMAT geophysical data
were badly affected by ground conditions, limiting its
reliability. GBU re-inverted the data with a 2D algorithm
which partially accounted for the difficulties. Reprocessed data indicated zones of high resistivity
coincident with main target areas such as Batu Mas. A 3D
IP survey over Lakuwahi was undertaken in 2010 and
extended in 2011. Data confirmed high resistivity was an
effective way to target zones of potentially economic
mineralisation. Modelling of data defined drilling targets
in all major outcropping prospects at Lakuwahi but has
limitations determining trends within the significant but
deeper and non-outcropping Perak Basin, possibly due to
conductive layers in the overburden.
Since 2008 GBU has drilled over 650 diamond
drillholes for just over 58,000m, using man-portable
drilling rigs and delineated a sizeable polymetallic
resource of 82Mt containing 1 Mozs of gold, 68 Mozs
silver and just over 1 Blbs of both lead and zinc (JORC
reporting standard). In 2014 a successful privatisation of
Robust Resources Limited has meant Lakuwahi is now a
majority-owned Indonesian project.Although current
metal prices and restrictions on the export of metal
concentrates from Indonesia have made it uneconomic to
develop the large polymetallic sulphide deposits, GBU are
pushing ahead with feasibility studies of a mining project
with conventional CIP treatment of oxide and transition
zones, over a 10 year mine life, producing 20,000 ozs Au
and 600,000 ozs Ag per annum.
From the onset of exploration GBU placed a priority on
engagement with the local communities on Romang,
enlisting the aid of Queensland Universitys Centre for
Social Responsibility in Mining for research and support
services with the companys community relations
strategies. The project now has a permanent staff of 6
CSR personnel which assist with daily running of the
exploration program and will help transition the
community into becoming involved with the proposed
mining project.
Fig. 2: Plan view of Romang Island showing magnetic data

from aerial survey.
GEOLOGY - LITHOLOGY
Romang consists of twin volcanic edifices at the
eastern end of the Sunda-Banda Arc where the Australian
Plate, moving northwards at 7cm per year, meets the
Eurasian Plate and becomes subducted beneath it, creating
island arc volcanism along its length. However, volcanic
activity in the Wetar Collision Zone, a 400 km long
segment extending from Alor to Romang Islands, ceased
between 8 - 3 Ma due to incorporation of less dense
continental crustal material from the Australian Plate into
the subduction zone. (Elburg et al, 2005). Herrington et al
(2011) observed that progressive incorporation of
continental crust into the Wetar Zone from collision with
the Australian Plate is coincident with formation of Aurich volcanogenic massive sulphide deposits on Wetar and
Romang.
Major rock types on Romang consist of an early
andesite-dacite sequence (lavas, pyroclastic agglomerates,
crystal/ lithic tuffs + intrusives) and a late daciticrhyodacitic succession, best exposed in north Romang. In
south Romang, basement andesites, known as Lakuwahi
Volcanics, are host to Lakuwahi mineralisation, they are
thought to consist of lavas, agglomerates and tuffs but
petrographic studies highlight the difficulty in identifying
original rock types due to intensity of hydrothermal
alteration which has invariably destroyed or overprinted
original texture and mineralogy. Volcanic activity was
submarine and although data is sparse, Lakuwahi
Volcanics, are thought to be between 1.4 2 Ma.
Lakuwahi Volcanics are unconformably overlain by
Upper Volcaniclastics, a series of lithic/lapilli tuffs, tuff
breccias, conglomerates. They are in turn overlain by Reef
Limestones varying from 1 to 20m in thickness containing
fossils such as corals and bivalves. The limestones are
characterised by karstic textures of varying degrees. No
work has been done yet on obtaining paleontological ages.
77
PROCEEDINGS
LWD067 0-101m: 101m @ 13.1% barite), sub-seafloor,

sub-horizontal breccias/stockwork or exhalatives.
Fluids also vented at the sea floor producing barite-rich
exhalatives (BEX) from 1-10m thick, similar to those
described at many VMS deposits around the world such as
Kuroko and Billitons mines on Wetar. However, in
contrast to Wetar where exhalatives overlie Cu-rich
massive sulphide lenses, Lakuwahi BEX horizons are
associated with Pb-Zn mineralisation with only minor Cu.
The best preserved feeder-zone at Lakuwahi is nonoutcropping in the Perak Basin, within a down-faulted
graben filled with lagoonal sediments protecting it from
erosion. High-grade Pb + Zn within the BEX layer
extends for 250m in a E-W direction with an average
interval of 2.1m @ 1.41 g/t Au + 266 g/t Ag + 30.1%
barite + 0.32% Cu + 5.6 %Pb + 7.2% Zn. The BEX
horizon proximal to breccia feeder zones (vent sites)
usually consist of chaotic debris with clasts of host
volcanics, re-worked BEX and even re-worked sulphides
mineralisation. Debris zones grade laterally for 100-200m
away from the vent sites into a more homogeneous barite
sand with abundant medium-coarse grained barite
crystals. In contrast with Wetar, the majority of Au/Ag +
base metal sulphides at Lakuwahi were precipitated in the
sub-seafloor environment as opposed to the seafloor
surface (Fig 2). However, although exhalatives are
volumetrically minor they are an important component of
ore reserves due to their high-grade precious and base
metal content. Breccia veins of high-grade base metals,
sub-parallel to feeder zones, have been intersected in some
deposits but with limited follow-up at this stage. They can
contain high amounts of base metals and offer significant
upside for future exploration work (eg LWD352 - 17.0m
at 25.67% Pb + Zn + Cu at Batu Mas Deeps).
Gold at Lakuwahi has only been recorded
petrographically as rare fine grains of electrum. Ag is
thought to occur within the galena lattice and also
associated with rare sulphosalts (tetrahedrite/ tennantite).
The main phase of polymetallic mineralisation retains
predominantly low-sulphidation characteristics with
galena-sphalerite-pyrite-chalcopyrite associated with silica
+/- sericite alteration. However there are patches of highsulphidation type ore which appear to be a later stage of
the main phase. Petrology has identified trace enargite
and, covellite in late fine-grained intergrowths with galena
and sphalerite.
A late low sulphidation style of mineralisation, likely
emplaced when the hydrothermal system was waning, is
present but spatially separate from the main polymetallic
deposits. Impressive colloform banded quartz with
carbonate replacement textures is widespread at Batu
Jagung and Kampung Lama contain very low Au and
variable to high Ag.
After cessation of the main sub-seafloor and exhalite
activity the Lakuwahi Volcanics were covered by
volcaniclastics. The top of Upper Volcaniclastics is
marked by a distinct reef limestone with coral and bivalve
fossils. Between 0.9 0.6Ma the hydrothermal system
was reactivated creating acidic Mn-rich fluids with highly
anomalous Ba, Zn, Pb, As, Sb and Tl. When these fluids
rose upwards along the breccia feeder zones they were
buffered as they encountered limestone and replaced the
carbonate-rich horizon.
There are at least two fault-bound basins (Perak Basin,

Hitam Basin) which contain Unconsolidated Sediments, a
series of shallow lagoonal sediments overlying Reef
Limestone. They comprise clays, mudstone and limestone
rubble/talus derived from adjacent uplifted reef. This
sequence can reach up to 40-50+ m thickness in the
basins.
The structural framework is typical of subductionrelated compressional settings including EW (arc-parallel)
accretionary faults, NS transfer faults and NW/NE
conjugate fractures. Analysis of modelled magnetic and
resistivity data has identified the importance of ring
structures to controlling the main Lakuwahi deposits. The
ring structures reflect caldera formation in a submarine
setting at the top of the Romang volcanic edifice.
MINERALISATION
The Lakuwahi hydrothermal system, driven by a
postulated deep-seated igneous intrusion, consists of
multiple structurally-controlled, linear to pipe-like, breccia
feeder zones, which brought mineralised fluids to seafloor
and sub-seafloor levels, depositing Au/Ag-rich
polymetallic mineralisation. The hydrothermal system has
been responsible for complex multiple mineralizing events
and it is still weakly active at several sulphur-rich
fumeroles.
The earliest phase is a typical low-sulphidation
assemblage spreading laterally from feeder structures
creating a zoned alteration pattern from inner silicasericite-pyrite with low gold tenor (0.1-0.3 g/t Au) and
little or no base metals, grading out to weak silica and clay
alteration and finally to a distal zone of propylitic
alteration. It is regarded as a ground preparation phase
creating bodies of hard, competent rock.
The second phase involved brittle re-activation along
feeder structures, creating breccias and stockwork zones
down to at least 150m below the seafloor. Rising
hydrothermal fluids deposited silica-barite with Au-AgPb-Zn-Cu-As-Sb. This second phase consisted of multiple
fluid and tectonic events with re-brecciation of breccias
and overprinting stockworks. Sulphide and sulphate
precipitation occurred along feeder zone breccias and
more importantly along sub-horizontal zones of breccia
and stockwork veining which extend out from feeder
structures and parallel to sub-parallel with the paleo
seafloor (Figure 2). Because the original lithology has
largely been obliterated it is difficult to surmise if these
zones represent lithological horizons such as lava flows
which were favorable to silica-sericite alteration and
therefore brittle deformation or they represent a depth at
which favorable PT conditions existed which promoted
precipitation of sulphides (eg by fluid boiling), in the subseafloor setting. Preliminary fluid inclusion work indicates
hydrothermal fluids were re-circulated seawater with
minimal or no magmatic input, Fluids were intermediate
temperature solutions (225 C -255 C) with an average
salinity of 3.2 wt% NaCl (OKane, 2013). Ar/Ar dating of
alteration assemblages give an age of 1.7 Ma.
Barite has an important relationship to polymetallic
mineralisation with Ba present at concentrations of 1-10%
Ba in all zones of economic Au/Ag mineralisation
regardless of whether in breccia feeder zones (eg
78
PROCEEDINGS
Fig. 2: Scematic section showing proposed depths of

mineralisation in Romang and Wetar producing differing
precipitation conditions accounting for the contrasting
dominance of PbZn v Cu sulphides.
The final and currently most economically important

event has been supergene alteration with formation of an
oxide zone above the water table. Au-Ag and Cu have
been re-distributed throughout the zone particularly with
Cu and Ag enriched at the water table. It should be
highlighted the zones of significant supergene enrichment
largely
correlate
with
zones
of pre-existing
brecciation/stockwork sulphide mineralisation with
minimal lateral movement away from those. Some of the
most spectacular precious metal intersections at Lakuwahi
have been from the oxide horizon at Batu Mas eg
LWD051 60m @ 5.06 g/t Au + 88 g/t Ag from surface.
`CONCLUSIONS
The Lakuwahi deposit on Romang Island is an
important new large VMS discovery in eastern Indonesia.
It has many similarities with nearby Wetar Island but also
significant difference such as dominance of Pb-Zn over
Cu and precipitation of majority of sulphides in subseafloor environment both of which suggest cooler
hydrothermal fluids were present and/or a shallow depth
of the system at Romang compared to Wetar (Figure 2).
This is supported by the presence of overlying reef
limestones at Romang (suggesting <200m depth) which
contrast with Wetar which is overlain by Globigerinabearing limestone which indicate a depth of up to 2km
(Scotney et al, 2005).
ACKNOWLEDGEMENTS
Many thanks to the Lakuwahi exploration teams for
their contributions to bring the exploration prospect at
Lakuwahi towards the reality of a producing gold-silver
mine.
REFERENCES
Elburg, M. A., 2005. Journal of Volcanology and Geothermal
Research, Volume 140, Issue 1-3, p. 25-47.
Hanington et al 1999, Economic Geology, 8, p. 324356
Herington et al, Gondwana Research, 19, Issue 3, 583593
OKane, L 2013, Honours Thesis University of Qld
Scotney et al., 2005 Mineralium Deposita 40: 7699
79
PROCEEDINGS
The metamorphic rock-hosted gold mineralization at Bombana (Southeast

Sulawesi) and Buru Island (Maluku): Their key features and significances for gold
exploration in Eastern Indonesia
Arifudin Idrus1*, Sukmandaru Prihatmoko2, Ernowo Harjanto3, Franz Michael Meyer4, Irzal Nur5, Wahyu Widodo3
& Lia Novelia Agung3
1
Department of Geological Engineering, Gadjah Mada University,Yogyakarta

2
PT. SJR-Pama Group, Jakarta.
3
Geological Agency of Indonesia, Bandung
4
Department of Mineralogy and Economic Geology, RWTH Aachen University, Germany
5
Department of Geological Engineering, Hasanuddin University, Makassar
Corresponding author: arifidrus@ugm.ac.id
ABSTRACT
In Indonesia, gold is commonly mined from epithermal-, porphyry-, and skarn-type deposits that are commonly found in
volcanic belts along island arcs or active continental margin settings. Numerous gold prospects, however, were recently
discovered in association with metamorphic rocks. This paper focuses on metamorphic rock-hosted gold mineralization in
Eastern Indonesia, in particular the Bombana (SE Sulawesi) and Buru Island (Maluku) prospects. At Bombana, goldbearing quartz-veins are hosted by the Pompangeo metamorphic complex. Sheared, segmented veins vary in thickness from
2 cm to 2 m. Gold is mainly present in the form of free gold among silicate minerals and closely related to cinnabar,
stibnite, tripuhyite, and in places, minor arsenopyrite. The gold distribution is erratic, however, ranging from below
detection limit up to 134 g/t. At least three generations of veins are identified. The first is parallel to the foliation, the
second crosscuts the first generation of veins as well as the foliation, and the late-stage laminated deformed quartz-calcite
vein represents the third mineralization stage. The early veins are mostly massive to crystalline, occasionally brecciated,
and sigmoidal, whereas the second-stage veins are narrower than the first ones and less subjected to brecciation. Gold
grades in the second- and third-stage veins are on average higher than that in the earlier veins. Microthermometric and
Raman spectrometric studies of fluid inclusions indicate abundant H2O-NaCl and minor H2O-NaCl- CO2 fluids.
Homogenization temperatures and salinities vary from 114 to 283 C and 0.35 to 9.08 wt.% NaCl eq., respectively. Crushleach analysis of fluid inclusions suggests that the halogen fluid chemistry is not identical to sea water, magmatic or
epithermal related fluids, but tends to be similar to fluids in mesothermal-type gold deposits.
In Buru Island (Gunung Botak and Gogorea prospects), two distinct generations of quartz veins are identified. Early quartz
veins are segmented, sigmoidal discontinuous and parallel to the foliation of the host rock. This generation of quartz veins
is characterized by crystalline relatively clear quartz, and weakly mineralized with low sulfide and gold contents. The
second type of quartz veins occurs within the mineralized zone of about 100 m in width and ~1,000 m in length. Gold
mineralization is intensely overprinted by argillic alteration. The mineralization-alteration zone is probably parallel to the
mica schist foliation and strongly controlled by N-S or NE-SW-trending structures. Gold-bearing quartz veins are
characterized by banded texture particularly following host rock foliation and sulphide banding, brecciated and rare
bladed-like texture. Alteration types consist of propylitic (chlorite, calcite, sericite), argillic and carbonation represented by
graphite banding and carbon flakes. Ore mineral comprises pyrite, native gold, pyrrhotite, and arsenopyrite. Cinnabar and
stibnite are present in association with gold. Ore chemistry indicates that 11 out of 15 samples yielded more than 1 g/t Au,
in which 6 of them graded in excess of 3 g/t Au. All high-grade samples are composed of limonite or partly contain limonitic
material. This suggests the process of supergene enrichment. Interestingly, most of the high-grade samples contain also
high concentrations of As (up to 991ppm), Sb (up to 885ppm), and Hg (up to 75ppm). Fluid inclusions in both quartz vein
types consist of 4 phases including L-rich, V-rich, L-V-rich and L1-L2-V (CO2)-rich phases. The mineralizing hydrothermal
fluid typically is CO2-rich, of moderate temperature (300-400 C), and low salinity (0.36 to 0.54 wt.% NaCl eq). Based on
those key features, gold mineralization in Bombana and Buru Island tends to meet the characteristics of orogenic,
mesothermal types of gold deposit. Metamorphic rock-hosted gold deposits could represent the new targets for gold
exploration particularly in Eastern Indonesia.
Keywords: Characteristics, orogenic gold, Bombana, Buru Island, Indonesia
.
currently gold is not only found in volcanic terrain, but
also many discoveries of placer (secondary) and primary
gold mineralization are genetically occurred in association
with metamorphic rocks, for instance, Awak Mas
mesothermal (Querubin & Walters, 2011), Poboya LSepithermal (Wajdi et al., 2011) and Rampi (North Luwu)
(Idrus et al., 2016). Gold-bearing quartz veins are also
recognized in Derewo metamorphic belt at northern and
northwestern part of Central Range Papua. Some
exploration reports categorized the Derewo metamorphic-
INTRODUCTION
During last few decades, in Indonesia gold has mostly
been extracted from volcanic-hosted hydrothermal
deposits, including LS epithermal type e.g. Pongkor in
West Java, Gosowong in Halmahera Island, HS
epithermal type e.g. Martabe (Sumatra), Cijulang (Jawa)
and Lanut, Doup (Sulawesi), skarn type e.g. Erstberg, Big
Gossan, Kucing Liar, Deep Ore Zone (DOZ) in Papua and
porphyry type e.g. in Grasberg (Papua), Batu Hijau
(Sumbawa Island) and Tombulilato (Sulawesi). However,
80
PROCEEDINGS
related quartz veins into mesothermal gold deposit type. In

Bombana (SE Sulawesi) (Fig.1), gold deposit is hosted by
metamorphic rocks of the Pompangeo Metamorphic
Complex (Mtpm). In addition, in this area placer gold is
also being mined from Mio-Pliocene sediments of
Langkowala Formation. This placer gold is interpreted to
be sourced from primary gold deposit hosted by the
Mtpm. Similarly, gold deposit is also discovered in Buru
Island (Gunung Botak and Gogorea), Maluku (Fig.1),
which is hosted by Wahlua Metamorphic Complex (Pzw).
Both Mtpm and Pzw are of Upper Carboniferous until
Lower Permian age (Simanjuntak et al. 1993;
Tjokrosapoetro et al., 1993).
Aachen University, Germany and Geotechnology

Research Centre, LIPI, Bandung. Raman spectrometry and
crush-leach analysis of fluid inclusion was done at
Georesources Research Center, Bandung and Leoben
University, Austria, respectively.
Deposit Geology
A.
Bombana Gold Deposit

The area is occupied by Mio-Pliocene Langkowala
Formation (Tmls) consisting of conglomerate and
sandstone. The Langkowala Formation is unconformably
underlain by Mesozoic metasediments and metamorphic
rocks (Pompangeo Complex, Mtpm) (Simanjuntak et al.,
1993). The metamorphic rocks consist of mica schist,
quartzite, glaucophane schist and chert. The
metasediments and metamorphic rocks occupy the
Mendoke and Rumbia Mountains. Mica schist and
metasediments particularly meta-sandstone and marble are
commonly characterized by the presence of quartz
veins/veinlets with various width up to 2 meters,
containing gold in some places. Interpretative E-Wtrending faults, which are relatively parallel to the
foliation attitude of the metamorphic rocks apparently acts
as mineralization-hosting shear zones that formed first
generation of quartz veins/reefs in the area. The NE
trending faults are thought to be the main control of the
formation of second generation of quartz veins (that crosscut the foliations). The regional geology of Bombana
(including Langkowala) and local geological map are
shown in Fig. 2.
As explained previously, the quartz veins are
predominantly hosted by metamorphic rocks particularly
mica schist. Mica schist is the predominant rock type in
the area. Some outcrops at Wumbubangka shows a
general foliation of N 300oE/60o. Some foliation
variations e.g. N80E/60-70 was reported (Prihatmoko et
al., 2010). Petrographic study of the mica schist indicates
that the rock is abundantly composed of muscovite,
chlorite and quartz with a small amount of actinolite,
albite, epidote, sericite and opaque minerals. Based on
those mineral assemblages, it is considered that the
metamorphic rock is categorized into green schist facies
(cf.Yardley, 1989). It is also important to note that the
majority of the metamorphogenic-related gold deposits
worldwide are hosted by greenschist facies (GebreMariam et al. (1995).
Bom
Fig.1 Location map of study areas i.e. Bombana (SE Sulawesi)

and Buru Island (Moluccas).
The genetic type of the Bombana and Buru Island gold

mineralization are still debatable. This paper is aimed to
discuss some key characteristics of the primary deposit
including host rock petrology, quartz vein texture and
structure, hydrothermal alteration, ore mineral and
chemistry and mineralizing hydrothermal fluid properties.
It is expected that the result would be important for a
better understanding of the genesis of the gold
mineralization, and would be useful in designing future
exploration strategy for gold deposits in Indonesia.
RESEARCH METHODS
This study has been carried out through several
approaches including desk study, fieldwork and sampling
for laboratory analyses. There is no previous detailed
study and publication that was focused specifically on the
primary gold mineralisation both in Bombana and Buru
Island. Therefore during the desk study only few
literatures related to Bombana placer gold can be
reviewed, e.g. Makkawaru & Kamrullah (2009) and
Surono & Tang (2009). Quartz vein samples were taken,
and geochemically analyzed by Fire Assay combined with
AAS, gravimetric fire assay (GA), cold vapour AAS (CV)
conducted in ITS laboratory, Jakarta and AAS laboratory
of Georesources Research Center, Bandung. Data of
quartz vein assays from Prihatmoko et al, (2010) that were
taken adjacent to the studied area are also incorporated
into discussion. A single ICP-MS analysis from Buru
Island sample was done by Prof. Victor Okrugin in
Kamchatka University, Russia. Mineral chemistry of
stibnite as a diagnostic ore mineral was also analyzed
using EPMA (Electron Probe Micro Analyzer) at RWTH
Aachen University. Fluid inclusion in various generations
of quartz veins was microthermometrically analyzed by
LINKAM THMS600 heating and freezing stage at RWTH
Fig. 2 Geological map of Langkowala area occupied by

Langkowala Formation (Tml) and unconformably overlain
81
PROCEEDINGS
Paleozoic metamorphic rocks (Pompangeo Complex; Mtpm) in

the south (Wumbubangka and Rumbia mountain range) (Base
map from Simandjuntak et al., 1993). Squared area indicates the
location area of this study.
The second quartz vein generation crosscut the first

generation quartz veins and as well as the foliation of
host-rocks (Fig. 4b); whereas the third vein generation is
characterized by deformed laminated quartz+calcite vein,
which is interpreted as the latest stage of vein formation in
the studied area (Fig. 4c).
In the Roko-Roko hill quartz veins (1-30 cm) hosted by
mica schist and metasediment are commonly massive to
crystalline quartz (druzy textures) with pseudomorph
bladed carbonate textures. Therefore, at least 2 later stages
of veinings (after the first generation veinings) could be
identified, including (1) vein breccias and (2) later quartz
veinlets, 1-10 mm, which are commonly crystalline and
containing native gold (Fig. 4f) (Prihatmoko et al., 2010).
Buru Island gold deposit

Similar to those of Bombana gold veins, gold
mineralization in Buru Island is also hosted by mica schist
of Carboniferrous to Permian Wahlua Metamorphic
Complex (Pzw) (Fig.3). Hence, it is important to note that
the rock characteristics and the ages of both Pzw and
Mtpm are exactly the same. Petrographic study exhibits
that mica schist in Buru Island is composed of muscovite,
chlorite and sericite suggesting of a green schist facies
(cf.Yardley, 1989).
Gogorea
Gn
Botak
Figure 4. Gold-bearing
orogenic
quartz
vein
characteristics:
Fig. 3 Geological map of Buru Island (Tjokrosapoetro et al.,
1993). Gunung Botak and Gogorea are occupied by Pzw
(Wahlua metamorphic rock complex).
(a). Brecciated/deformed quartz vein (first generation) which

is paralel to the foliation of the mica schist (N 300E/60), (b)
highly oxidized/mineralized deformed second quartz vein
cross cutting foliation, (c) A cluster of deformed laminated
quartz veins hosted by metasediment,
Gold mineralization in Buru occurred in form of quartz

veins/veinlets/reef. Two types/generations of quartz veins
are recorded namely (1) Quartz veins which are
segmented, sigmoidal, discontinous and parallel to the
foliation of the metamorphic rocks. The vein distribution
and pattern is intimately controlled by foliation orientation
in the area. Mineralogically, the quartz vein is lack of
sulfides, weak mineralized, crystalline, relatively clear and
gold may be poor; (2) Quartz veins occured within a
mineralized zone of about 100 m in width and ~1,000 m
in length. Gold mineralization is strongly overprinted with
argillic alteration zone. Although it is still lack of field
data, the mineralization-alteration zone is probably
parallel to the mica schist foliation. According to field
data and Buru geological map (cf. Tjokrosapoetro et al.,
1993), it is interpreted that gold mineralization may be
strongly controlled by N-S or NE-SW-trending geological
structures (strike-slip faults?). Artisanal and small scale
gold mining (ASGM) activities are currently concentrated
along the structural-controlled mineralization zone.
Buru Island
Field and hand specimen observation indicates that
gold-bearing quartz veins are characterized by vuggy,
banded texture particularly colloform following host rock
foliation and sulphide banding (Fig. 5a) and brecciated
texture. Bladed-like texture is also observed, but it is rare
(Fig. 5b). Those textures are more like developed in
classic LS epithermal vein deposits.
Fig. 5 (a) Handspecimen of second quartz vein type with

banding (colloform texture quartz vein following foliation),
graphite and sulphide banding, (b) Handspecimen of highly
oxidized/limonitic quartz vein with bladed-like texture indicating
a boiling condition (?).
GOLD-BEARING QUARTZ VEINS

Bombana
At least there are three generations of the quartz veins
identified. The first generation of quartz vein is parallel to
the foliation of mica schist, phyllite and metasediment
with general orientation of N 300oE/60o (Fig. 4a). It was
occasionally observed that this first generation of quartz
vein is crosscut by quartz veinlets/ stockwork/stringers.
However, a few anomalies from shallow gold

systems in the Yilgarn block of Western Australia are
notable. Comb, cockade, crustiform and colloform
82
PROCEEDINGS
cave (a natural cave) in Wumbubangka at the northern

part of the Rumbia mountain shows a high Au grade of
134 g/t. Few of the typical pathfinder minerals associated
with orogenic/metamorphic-hosted gold deposit are
stibnite and tripuhyite. The chemical composition of the
minerals analyzed using EPMA (Electron Probe Micro
Analyzer) indicates that both antimony-bearing minerals
(stibnite and tripuhyite) contain a significant amount of As
of up to 1 wt.%.
textures at the Racetrack deposit, Australia, deposited

from CO2-poor fluids in lower greenschist facies rocks are
also recognized (Gebre-Mariam et al., 1993). Similar
textures at the Wiluna gold deposits in subgreenschist
facies rocks, as well as 18Oquartz measurements as light as
67 per ml, provide some of the strongest evidence of
meteoric water involvement in some of the mesothermal
hydrothermal systems (Hagemann et al., 1992, 1994).
Although it is uncommon, but pseudomorph bladed
carbonate texture could be present in orogenic quartz
veins/reefs if the hydrothermal fluids forming the ore
deposit have the right phase separation condition
(personal comm.., Richard J. Goldfarb, 2011).
ALTERATION AND ORE MINERALOGY
Bombana
The wallrocks (metamorphic rocks) are strongly
weathered, so it is very rare to observe good outcrops in
the area. Trenching program by the company along the
spurs of Wumbubangka metamorphic mountain range has
opened up the soil cover, and exposed clearly the presence
of quartz veins and hydrothermally altered rocks. The
hydrothermal alteration types recognized in the field
includes silicification, clay-sericite-silica (argillic),
carbonate alteration and carbonization. Silicification is
represented by silicified metasediment and mica schist,
whereas clay-sericitesilica (argillic) is mostly present
surrounding quartz veins or along structural zones.
Prihatmoko et al. (2010) also reported the presence of
narrow clay-sericite alteration halo (tens cm to 1 m)
around the quartz veins in the Roko-Roko hill. Carbonate
alteration is typified by the presence of calcite
veinlets/stringers, while carbonization is represented by
rare occurrences of graphite/carbon with common black
color in the quartz vein/adjacent to the altered wall rocks.
The carbonization is considered to be one of the alteration
type
characteristics,
associated
with
orogenic/metamorphic-hosted gold deposit.
Quartz veins/reefs/veinlets contain very small amount
of sulphide minerals (up to 5 %). Pyrite, chalcopyrite,
cinnabar (HgS), stibnite (Sb2S3), tripuhyite (FeSbO4) and
rare arsenopyrite (FeAsS2) are present in the quartz veins
and silicified metamorphic wallrocks. Cinnabar is
typically pinkish red in color and present abundantly in
both primary mineralization and in placer gold deposit. On
the other hand, in the primary mineralization, cinnabar
commonly occurred in the form of mineralized layers
along foliations of the metamorphic rocks (Fig. 6a).
Stibnite and tripuhyite seem to be filling fractures parallel
to foliations (Fig. 6b) and disseminated within the
silicified wall rocks. In general, gold is very fine-grain,
but occasionally native gold is visible in quartz veins (Fig.
6c,d).
Cinnabar and stibnite are genetically closely related to
gold mineralization. Those sulfides could be pathfinder
minerals for the exploration of the metamorphic-hosted
gold deposit. AAS ore chemistry indicates a very broad
and erratic variation of gold grade ranging from below
detection limit (0.005 g/t) to 84 g/t Au (based on present
study and Prihatmoko et al. (2010)), even a single analysis
of quartz vein sample (BVAL-01) from the Valentino
Fig. 6 Diagnostic sulfides associated with Bombana orogenic

gold mineralization: (a). Layer-like pinkish cinnabar paralel to
mica schist foliation, (b). Fibrous stibnite mainly parallel to the
foliations, (c). Visible native gold in multiple quartz veins, and
(d). Photomicrograph of free gold in quartz vein.
Buru Island
As outlined above, gold mineralization zone is
intimately associated with argillic-altered mica schist
delineating an obvious high Au grade zone of about 100 m
width and 1,000 m length. Clay mineral types
characterizing argillic alteration zone are unknown.
Petrographic analysis shows host rock is also
propyllitically altered typified by the presence of chlorite,
calcite and sericite. Carbonation alteration style
represented by graphite banding and carbon flakes is a
typical alteration type occurred in metamorphic-related
hydrothermal ore deposits. Ore mineralization is
characterized by pyrite, native gold, pyrrhotite and
arsenopyrite. As found in Bombana, cinnabar and stibnite
are also identified in association with gold. In general,
sulphide minerals are rare (<3%). This is consistent with
mineralogical features of other metamorphic rock-hosted
gold mineralizations worldwide (cf. Groves et al., 1998,
2003).
ORE MINERALIZING FLUIDS
Bombana
A total of 6 quartz veins/reefs from three different
generations were prepared for fluid inclusion analysis.
This study has enabled to understand the characteristics
including temperature, salinity and composition of
mineralizing hydrothermal fluids that formed the three
generations of quartz veins. The data show that Tm of
fluid inclusions hosted by first generation of quartz veins
(that are parallel to the foliation) tend to be lower ranging
from -2.3 to -10 C (mean -3.2 to -5.9 C) corresponding
to relatively higher salinity ranging from 5.26 to 9.08
wt.% NaCl eq.) in comparison to those of other
83
PROCEEDINGS
generations of quartz veins/reefs. The temperature of

homogenization (Th), interpreted to be the formation
temperature of the first generation of quartz vein varies
from 185 to 245 C, that are relatively higher than those of
other two generations of quartz veins/reefs. The second
generation of quartz veins, that cross-cut foliation and
have generally higher gold content, is formed in moderate
temperatures of 132-283 C (mean 158-209 C) and
salinity of 3.55-5.86 wt.% NCl eq. The latest generation
stage of veining represented by quartz+calcite laminated
veins was originated at the lowest temperature of 114-176
C and salinity of 0.35-4.03 wt.% NaCl eq.
Fig. 8 Fluid inclusion study: (a) and (b). CO2-rich L-V fluid
inclusions hosted by quartz veins, and (c). Raman spectrometric
analysis of carbonic fluid inclusion containing dissolved CO2
with certainty up to 92.73%.
Fig.7 Temperature of homogenization (Th) vs salinity of fluid
inclusions from three different quartz vein generations at
Bombana metamorphic-hosted gold deposit. The hydrothermal
fluid evolution of the three types of quartz veins are also shown
and discussed in the text. Schematic model of fluid evolution is
adapted from Shepherd et al. (1985).
Buru Island
A total of 5 quartz veins/reefs from two different types
were analysed for fluid inclusion study. Four samples
contain measurable fluid inclusions, and those of one
sample (B05VA) are too small to be measured. The data
in Table 1 show that Tm of fluid inclusions hosted by first
type of quartz veins (that are crystalline, clear, weak
mineralized and parallel to the foliation) tend to have Tm
ranging from -0.1 to -0.3 C (average -0.22 C)
corresponding to salinity ranging from 0.18 to 0.53 wt.%
NaCl eq.(average 0.36 wt.% NaCl eq.), relatively lower
than those of second quartz vein type (Tm = -0.2 to 0.3
C; average -0.27 C) which correspond to salinities of
0.36 to 0.54 wt.% NaCl eq., averaging 0.48 wt.% NaCl eq.
The temperature of homogenization (Th), interpreted to be
the formation temperature of the first type of quartz vein
varies from 234 to 354 C, that are relatively lower than
those of second quartz veins type (Th = 321 to 400 C).
Petrographic study indicates that fluid inclusions in
both quartz vein types consist of 4 phases including Lrich, V-rich, L-V-rich and L1-L2-V (CO2)-rich phases. In
addtion, Sample B05VB is characterized by abundant Vrich and L-rich inclusions which may imply a boiling
condition with an elevated temperature of 400 C. In fact,
this sample was taken from Gunung Botak where the
artisanal and smal-scale mining (ASGM) situated. Crushleach analysis of fluid inclusions from Bombana and Buru
Island gold veins suggests that the halogen fluid chemistry
(Br/Cl vs I/Cl plot) is not identical to magmatic or
epithermal related fluids, but tends to be similar to fluids
in mesothermal-type gold deposits (Fig. 9).
Figure 7 displays the plotting between Th and salinity

of fluid inclusions from all quartz vein generations. It is
clearly indicatived that the first quartz vein generation
underwent an isothermal mixing with fluids with
contrasting salinity, and is interpreted that the first quartz
generation is dominantly originated from hydrothermal
magmatic fluid mixing with metamorphic fluids. During
the mixing, the temperature change is minor or relatively
isothermal, but the salinity decreases significantly. The
second and third quartz vein generations are likely formed
from mixing of the magmatic and metamorphic fluids, and
with cooler less saline meteoric water. This is shown by a
systematic decrease of temperature and salinity (Fig. 7).
The evidences of the contribution of metamorphic fluid,
hydrothermal magmatic fluids and meteoric water that
formed the quartz veins are represented by H2O-NaClCO2 fluid inclusions (Fig. 8a,b). Petrographically the
carbonic fluid inclusions are occasionally observed and
may contain small portion of CO2, probably max. 4%
CO2 (personal comm., Richard J. Goldfarb, 2011). Raman
spectrometric analysis confirms the presence of dissolved
carbon dioxide (CO2) in primary fluid inclusion with a
certainty of up to 92.73% (Fig. 8c).
Table 1 Microtermometric data of fluid inclusions within quartz

veins associated with gold mineralization in Buru Island
84
PROCEEDINGS
-2
log (Br/Cl)
-2.5
Sea water
-3
Brusson lode gold

AM_Unmin.
AM_Min.
-3.5
QV_ WB
Capitan
Magmatic
QV_KB
QV_B
-4
-7
-6
-5
-4
-3
-2
log (I/Cl)
0.10000
b
Mesothermal gold
0.01000
Br/Cl
Epithermal Au
0.00100
AM_Unmin.
Sea water
AM_Min.
QV_ WB
QV_KB
Porphyry Cu
QV_B
0.00010
0.000000
0.000001
0.00001
0.0001
0.001
0.01
I/Cl
Fig. 9 Crush-leach analysis of halogen content (I/Cl and Br/Cl
ratios) in fluid inclusion showing mineralizing fluids are not
identical to magmatic fluid (a), epithermal (meteoric waterdominated) and porphyry Cu (b), but tends to be similar to fluids
in mesothermal-type gold deposits.
CONCLUSIONS AND SIGNIFICANCES FOR

EXPLORATION
Conclusions
Briefly concluded that primary gold mineralization in
Bombana particularly in Wumbubangka area, at the
northern flank of the Rumbia mountain range is
predominantly hosted by mica schist which is
petrologically categorized into greenschist facies. This
type of metamorphic facies mostly hosts the orogenic gold
deposits worldwide, e.g. Mt. Charlotte, Lancefield and
Golden Mile (Gebre-Mariam et al., 1995; Goldfarb, 2009).
The presence of pathfinder minerals such as cinnabar,
stibnite and tripuhyite genetically indicates that the
orogenic gold deposit in the Bombana is emplaced into
transition between epizonal and mesozonal referred to the
conceptual model of orogenic gold deposit (cf. Groves et
al., 1998, 2003) (Fig. 10). It implies that the
mineralization may be formed at approximately 5 km
depth below paleosurface.
In addition, the observable characteristics of goldbearing quartz veins/veinlets have met with the criteria of
orogenic gold type, i.e. sheared/deformed, segmented,
brecciated and occasionally sigmoidal, which are the key
indications for brittle condition of the epizonal-mesozonal
transition. The quartz veins/reefs are commonly
characterized by massive and crystalline textures.
However, druzy and pseudomorph bladed carbonate
textures are also occasionally recognized. Although it is
uncommon, but bladed carbonate could be present in
orogenic quartz veins/reefs if the hydrothermal fluids
forming the deposit have the right phase separation
situation (personal comm., Richard J. Goldfarb, 2011).
Ore mineralizing fluid is characterized by moderate to low
salinity ranging from 5.26 to 9.08 wt.% NaCl eq., 3.555.86 wt.% NCl eq. and 0.35-4.03 wt.% NaCl eq. as well as
moderate to low temperature of homogenization (Th)
varying from 185 to 245 C, 132 to 283 C, and 114 to 176
C for first, second and third generation veins,
respectively. CO2-rich fluid inclusion is present in small
portion, but its presence is well confirmed by Raman
spectrometric data. Crush-leach analysis of fluid inclusion
shows that mineralizing fluid characteristics are not
identical to both epithermal (meteoric water-dominated)
85
PROCEEDINGS
and magmatic fluids, but tends to fit with the properties of

mesothermal related fluids.
Significance for Gold Exploration

Although gold mineralization both in Bombana and
Buru Island has some discrepancies, but by considering all
key features discussed above, the primary metamorphichosted gold mineralization type tends to meet the criteria
of orogenic gold type (cf. Groves et al., 1998; 2003),
rather than epithermal or other hydrothermal deposit
types. Therefore, the discovery of the metamorphic-hosted
gold deposit in the Rumbia metamorphic mountain range,
Buru Island (Gunung Botak and Gogorea prospects) and
its vicinity has opened up more targets and challenges for
gold exploration in the region, and other terrains in
Indonesia particularly Eastern Indonesia that have
identical geological setting. It is also important to note
that all high-grade samples are originally or containing
limonitic/oxidized materials, that suggest the role of
supergene enrichment. A further study on other genetic
parameters of the deposit such as ratios of detailed
mineralogy (sulfides & gangue), base metals, pathfinder
elements, specific features, etc is needed for a better
understanding of the genesis of the Bombana and Buru
Island gold deposits.
Fig. 10 The Bombana and Buru Island metamorphic-hosted gold

deposit plotted on the conceptual orogenic gold deposit model
from Groves et al. (1998, 2003) emplaced into shallow level at
the transition between epizonal and mesozonal (approximately 5
km below paleosurface).
ACKNOWLEDGEMENTS
The authors wish to express a gratitude to the
Directorate of Higher Education, Department of National
Education, Indonesia, for International Competitive
Grant
2009
with
contract
number
of
694/SP2H/PP/DP2M/X/2009 granted to the first author as
a principal researcher. We are also indebted to the Energy
and Mineral Resources Agency of Southeast Sulawesi and
Bombana Regency, respectively for their permission. The
supports and permission from Management of PT. Panca
Logam Makmur are much acknowledged. Some field
data/ report provided by PT. AGC Indonesia are
acknowledged. This study is also made possible through
financial support from Barry Smith and Research
Cooperation Program 2013 between the first author with
Geological Resources Research Center (PSDG) Bandung.
Prof. Victor Okrugin (Kamchatka University, Russia)
provided an ICP-MS single analysis. Crush-leach analysis
of fluid inclusion was done in Leoben University, Austria.
Those supports are highly acknowledged. Fadlin Idrus and
Satriadin Abdullah are thankful for their fieldwork
assistances and data processing.
Similarly, Buru Island gold mineralization is hosted by

mica schist, which is composed of muscovite, chlorite and
sericite, thus this metamorphic rock is grouped into green
schist facies. Two generations of gold-bearing quartz
veins are identified. First quartz veins are typically
segmented, sigmoidal, discontinous and parallel to the
foliation of the metamorphic rocks. Second quartz veins
occurred within a mineralized zone of about 100 m in
width and ~1,000 m in length. The mineralized quartz
vein is probably parallel to the mica schist foliation.
Mineralized zone is generally brecciated and overprinting
with argillic alteration zone with N-S or NE-SW
orientation. Mineralized zone may strongly be controlled
by N-S or NE-SW-trending strike-slip faults. Second
quartz vein texture is characterized by brecciated, banding
texture such as colloform following foliation, sulphide
banding and occasionally bladed-like texture. Host rock is
altered to propylitic, argillic, silicification and
carbonation. Carbonation is shown by graphite banding
and carbon flakes associated with quartz banding.
Typically, sulphide minerals are rare (<3%). Cinnabar,
stibnite, arsenopyrite, pyrite and native gold are identified,
those mineral are indicative of mineralization systems of
orogenic gold. Assay results indicate that 11 of 15
samples yielded more than 1 g/t Au, in which 6 of them
are in excess of 3 g/t Au. It can be noted that all highgrade samples are originally or containing limonitic
materials, that suggest the role of supergene enrichment.
Interestingly, most of the high-grade samples contain also
high grade As (up to 991ppm), Sb (up to 885ppm), and Hg
(up to 75ppm). Hydrothermal fluid is typified by CO2-rich
fluid, moderate temperature of 300-400 C (mesothermal
T range) and a typical low salinity (0.36 to 0.54 wt.%
NaCl eq), which suggests that the metamorphic fluid is
responsible for the formation of the Buru gold deposit.
REFERENCES
Gebre-Mariam, M., Hagemann, S. G., and Groves, D. I., 1995, A
classification scheme for epigenetic Archaean lode-gold
deposits. Mineralium Deposita 30: 408-410.
Groves, D. I., Goldfarb, R. J., Gebre-Mariam, M., Hagemann, S.
G. and Robert, F., 1998, Orogenic gold deposit: A proposed
classification in the context or their crustal distribution and
relationship to other gold deposit types. Ore Geology
Review 13: 7-27.
Groves, D. I., Goldfarb, R. J., and Robert, F., 2003, Gold deposit
in metamorphic belts: Overview or current understanding,
outstanding problems, future research, and exploration
significance. Economic Geology 98: 1-29.
Hagemann, S.G., Groves, D.I., Ridley, J.R., Vearncome, J.R.,
1992. The Archaean lode-gold deposits at Wiluna, Western
Australia. High level brittle-style mineralisation in a strikeslip regime. Econ. Geol. 87, 10221053.
86
PROCEEDINGS
Hagemann, S.G., Gebre-Mariam, M., Groves, D.L., 1994.

Surface-water influx in shallow-level Archean lode-gold
deposits in Western Australia. Geology 22, 10671070.
Idrus, A., Mansur, S., Ahmad, Rahmayuddin & Mahdi, A., 2016,
Occurrences and Characteristics of Gold Mineralization in
Rampi Block Prospect, North Luwu Regency, South
Sulawesi Province, Indonesia, Jounal of Applied Geology,
Volume 8, January-June 2016 Edition, accepted.
Prihatmoko, S., Lubis, H., Hernawan, S., 2010, Evaluation report
of Bombana gold prospects, Southeast Sulawesi,
unpublished report, PT. AGC Indonesia, 41p.
Querubin, C.D., Walters, S., 2011, Geology and Mineralization
of Awak Mas: A Sedimentary Hosted Gold Deposit, South
Sulawesi, Indonesia, Proceedings of The Sulawesi Mineral
Seminar, Manado 28-29 November 2011, p. 211-229.
Simandjuntak, T.O., Surono, Sukido, 1993, Peta geologi lembar

Kolaka, Sulawesi, P3G, Bandung.
Shepherd, T.J., Rankin, A.H., Alderton, D.H.M., 1985, A
practical guide to fluid inclusion, Blackie, London, 239p.
Tjokrosapoetra, S., Budhitresna, T., Rusmana, E., 1993, Peta
Geologi Lembar Buru, Maluku, Pusat Penelitian and
Pengembangan Geologi (P3G), Bandung.
Wajdi, M.F., Santoso, S.T.J., Kusumanto, D., Digdowirogo, S.,
2011, Metamorphic Hosted Low Sulphidation Epithermal
Gold System at Poboya, Central Sulawesi: A General
Descriptive Review, Proceedings of The Sulawesi Mineral
Seminar, Manado 28-29 November 2011, p. 201-210.
Yardley, B. W. D., 1989, An introduction to metamorphic
petrology. Longman Scientific & Technical, Essex, 247p.
87
PROCEEDINGS
Lesson Learnt from Nickel Sulfide Exploration in Latao Block, Southeast Sulawesi:
A Geological and Geometallurgical Perspective
Budhi Kumarawarman, Gde Handojo Tutuko, and Moh. Erwin Syam Noor
PT Vale Indonesia
ABSTRACT
A comprehensive nickel sulfide exploration had been undertaken in Latao between years 2008 to 2012. The nickel sulfide
mineralization in Latao is not primary magmatic origin. It is more low-grade disseminated nickel sulfide type, similar with
the nickel sulfide deposit in Dumont (Canada) and Ronnbacken (Sweden) which typically large in tonnage but low in nickel
grade. The ore bodies were predicted to be large. Unfortunately, result of metallurgical testwork was far below expectation.
If the Ronnbackens samples were reported to be able to produce concentrate with a grade of 28% at 80% recovery, no
nickel upgrading was resulted from Lataos samples in the metallurgical testwork. The primary reason was because not all
of the Lataos pentlandites are able to liberate. Despite Latao has similar nickel grade with the Ronnbackens, pentlandite
grains of Latao are smaller than the Ronnbackens. Moreover, these fine pentlandite grains are found mostly as inclusion in
serpentines, or locked by the serpentine. Mineral examination on feed samples of the flotation test also suggested highly
nickel oxidized serpentine rock. The trace amounts of nickel sulfide minerals that are present are secondary alteration
products from primary nickel sulfide (pentlandite) of which no longer exist in its original form of chemical composition.
This highly oxidized ore does not respond well to sulfide flotation.
Overall, Latao is an example of new-frontier for nickel exploration in Indonesia. Aside to Latao, PT Vale Indonesia, Tbk
also found trace of nickel sulfide occurrence in Sorowako, South Sulawesi. These two findings show that it is likely possible
to find another type of nickel sulfide deposits in Indonesia. Nevertheless, a lesson learnt can be taken from Latao
exploration result. It is suggested that exploration shall not only focus on the nickel grade, but the nature of the sulfide
minerals shall also be the primary consideration when exploring a nickel sulfide deposit. Geological structures also play
important role in mineralization processes, particularly when dealing with the non-primary magmatic nickel sulfide deposit.
Thus, detail structural geology study must be carried out as well to recognize the position, orientation and dipping of the
hydrothermal fluids channels. These parameters are important to define the proper exploration method, including the
sample handling and sample preparation method.
Keywords: Latao, nickel sulfide, metallurgical testwork, grain size, locking, liberation
INTRODUCTION
Latao is located in the Batuputih subdistrict of North
Kolaka Regency. It is situated approximately 150 km
southwest of Sorowako, or about 2.5 hours drive from
Sorowako (Figure 1). During 2008 2012, a
comprehensive nickel sulfide exploration was carried out
in this area by PT Vale Indonesia, Tbk (PTVI), or
formerly known as PT INCO, Tbk. Totally, 53 diamond
drillholes of total length 7,911.8 meters were drilled in
Latao to confirm the nickel sulfide mineralization body,
supported with ground-magnetic and IP-resistivity data.
Nickel sulfide minerals were recognized from the core
samples, mostly occurring as fine-grained millerites and
pentlandites.
To obtain information regarding the geometallurgical
properties of the deposit, a series of metallurgical testwork
was also performed in 2012 and 2013. Partial extractions
were done for Lataos core samples using brominemethanol dissolution method to figure the amount of
sulfide nickel and silicate nickel proportion of respective
samples. Moreover, series of flotation tests were also
conducted to understand the geometallurgical properties
of the ores. The flotation tests were subsequently followed
by mineral locking and liberation analysis to further
recognize the mineralogical characteristics that may
influence the mineralogical testwork result.
Fig. 1 : Location and accessibility of Latao Block
NICKEL SULFIDE OCCURRENCE

Generally, nickel sulfide occurs as magmatic deposit.
Foose et al (1995) explained that these magmatic sulfide
deposits are sulfide minerals concentration in mafic and
ultramafic rocks derived from immiscible sulfide liquids.
Exsolution of immiscible sulfide liquid from mafic to
ultramafic magmas is the fundamental processes that from
the magmatic sulfide deposit. Once formed, droplets of
88
PROCEEDINGS
immiscible sulfide liquid settle through less dense silicate

magma. The sulfide liquid acts as a collector for cobalt,
copper, nickel and platinum-group element (PGE) because
these elements are preferentially concentrated in sulfide
liquids at levels 10 to 100,000 times those in silicate
liquids. To a lesser degree, iron is also preferentially
partitioned into the sulfide liquid, and because of its
greater abundance, most immiscible sulfide liquid is iron
rich. The combination of physically concentrating dense
sulfide liquid and chemically concentrating elements in
the sulfide liquid is responsible for most economically
mineable magmatic sulfide deposit. The model of ore
profile succession in magmatic sulfide deposits is
illustrated in Figure 2. Furthermore, Figure 3 illustrates a
scheme that shows the difference between enrichment
process of nickel in laterite and sulfide deposits as
provided by Kadarusman (2008).
magmatic origin. Normal magmatic sulfides in ultramafic

system have Ni/Cu ratio ~20, while the Ni/Cu ratios of
Latao are ~5000.
Nickel sulfide mineralization in Latao were mostly
occurring in the form of fine-grained millerites and
pentlandites, although coarse-grained nickel sulfide
minerals do exist as well. They present disseminated in
the highly-serpentinised ultramafics to serpentinites and
altered tectonic breccia (Figures 6 and 7). Silica ribs and
magnetite rock bodies (Figure 8) are also found as
association with the deposit. In addition to the nickel
sulfide mineralization, PGE (Platinum Group Elements)
anomaly was also reported to occur in the residual soils.
Several samples of soil were containing anomalous
palladium concentration up to 51 ppb.
Geologically, Latao is a complex of ultramafic block
that is bordered by fault complex. Rusmana et al (1993)
interpreted the ultramafic block of Latao had undergone
thrusting onto the adjacent metamorphosed rocks of
Meluhu Formation (Figures 4 and 5). The thrust fault is
west-dipping and extending relatively north-south in the
eastern part of the Latao Block.
Fig.2 A) Schematic section of a magmatic sulfide deposit

showing the vertical gradation downward from disseminated to
massive ore (Foose et al, 1995)
Fig. 4 Regional geology map (Rusmana et al, 1993)
The nickel sulfide occurrence in Latao is likely

controlled by a hydrothermal alteration. The hydrothermal
fluids are possibly released through the channeling of
major structures networks and altering the ultramafic
rocks of Latao. The origin of hydrothermal fluids are
remain unknown. However, several authors (Hall, 1996;
Kadarusman et al, 2004) have recognized a Tertiary
westward movement subduction zone in the east of
Sulawesi that was subsequently followed by the collision
of Banggai-Sula and Tukang Besi microcontinents with
the Sulawesi arc of Sundaland margin. Kadarusman et al
(2004) interpreted that the subduction zone occuring
between 10 20 Ma (Miocene) had formed the NVC
(Neogene volcaniclastics). Aside to Kadarusmans
interpretation, Hinchberger et al (2004) had also
recognized a rifting in North Banda Sea Basin (NBSB)
that initiated westward subduction of NBSB oceanic
lithosphere beneath the Sulawesi at Tolo Thrust, starting
from 3 Ma (Pliocene) until today. Which subduction event
resulting in the magmatism that release hydrothermal
fluids for the alteration of Lataos rocks is still not clear,
but a hot-spring located in Purehu Village ( 15 km in the
east of Latao) provides a strong evidence of recent
magmatism activity in the neighborhood of Latao.
Fig.3 Schematic occurrence of nickel laterite development and

nickel sulfide occurrence from ultramafic rocks (Kadarusman,
2008)
GEOLOGY OF LATAO
Despite most nickel sulfide occurs as magmatic
deposit, nickel sulfide occurrence in Latao is somewhat
different. The deposit is more low-grade disseminated
nickel sulfide type, similar with the nickel sulfide deposits
in Dumont (Canada) and Ronnbacken (Sweden). Like the
Dumont and Ronnbacken (Royal Nickel, 2012; SRK
Consulting, 2011), nickel grades in Latao are about 0.2%
with amount of sulfide nickel around 0.1%. No massive
sulfides are found in Latao. Based on the ratios of Ni/Cu
in Latao, Lightfoot and Vandenburg (2013) wrote that
nickel sulfide mineralization in Latao is unlikely to have
89
PROCEEDINGS
through the supply of water, S and CO2. The reaction can

be summarized as follow (SRK Consulting, 2011):
Olivine (Ni) + H2O + S2 + CO2 Serpentinite + Brucite +
Fe3O4 + NixSy* + H2
*Ni rich sulfides
Fig.5 Re-interpretation of Latao geology based on mapping,

drilling and IP-Resistivity data (modified from Rafianto et al,
2012)
Fig. 6 :Core showing sulfide minerals disseminated in

serpentinite
The magnetite rock bodies occurrences were perhaps

also related with the serpentinisation processes and
hydrothermal fluids activity. Considering that they are
outcropped along the NNE-SSW Laburino fault zone, it is
likely possible that the magnetite rocks occurrence are
fault controlled. To explain the magnetite occurrence and
its relation to the hydrothermal alteration, we can refer to
occurrence of magnetite veins in Bou-Azzer ophiolite,
Morocco. Gahlan et al (2006) explained that the magnetite
vein found Bou-Azzers serpentinized harzburgite are
considered to be formed along the serpentinization
process. The source of iron was internal, supplied from
olivine. The mineralizing hydrothermal fluids utilized the
fractures for transportation and precipitation of iron,
which transported in ferrous hydroxide. These ferrous
hydroxides are then precipitated in the cracks and
fractures. This explanation is likely the plausible answer
on the question regarding the occurrence of magnetite
rock bodies.
Initially, the Latao mineralization was interpreted to
closely related with the combination of serpentinisation
process, listwaenite-birbirite silica alteration and
magnetite hematite (iron oxides) alteration (Rafianto et
al, 2012). However, further geochemical and petrological
study by Lightfoot and Vandenburg (2013) suggested that
Lataos rocks have low C (0.05w%) and low As (0.1ppm).
Therefore, a listwaenite style of alteration of the protolith
with hydrothermal nickel mineralization appears to be less
likely.
The pattern of the altered rocks in Latao (serpentinised

ultramafics, altered tectonic breccia, magnetite rock
bodies) show tendency of NNE-SSW orientation along the
NNE-SSE faults (Figure 5). In the eastern part of the
block, the altered serpentinised ultramafics diminish along
a NNE-SSW structures. The magnetite rock bodies are
also scattering along the NNE-SSW structures of the
Laburino fault that create the valley and the gulf of
Laburino and extending to Rokka bay in the north. Zones
of high-resisitivity and high-chargeability that suggest the
mineralization zone occurs somewhere between those
structures as well. IP-resistivity survey results indicate
these NNE-SSW structures are dipping east.
In Lelewawo village, the northeastern part of the block,
these east dipping NNE-SSW structures are intercepted by
the west-dipping thrust fault that emplaced the ophiolitic
rocks onto the phyllite and meta-limestone members of
the Meluhu Formation (Figure 5). It is remain unclear
whether the sulfide mineralization in the ultramafic body
occurred before or after the thrusting period. If the
mineralization took place prior to the thrusting event, then
the east-dipping NNW-SSE may serve as the main
channel for the hydrothermal fluids. And then, the
mineralized ultramafic rocks body was displaced by the
thrust fault onto the metamorphic rocks of Meluhu
Formation. On the contrary, if mineralization happened
after the thrusting event, then the main hydrothermal
fluids channel will be probably the west-dipping thrust
fault. In this case, the east dipping structures served as the
sub-channel that distribute the hydrothermal fluids to the
ultramafic rock body.
Along the alteration processes, serpentinisation of
ultramafic rock body occurred. Nickel was originally
located mainly in the olivine lattice within the ultramafic
rocks, such as dunites and peridotites (Rafianto et al,
2012). Due to serpentinisation of the olivine, the nickel in
the olivine was released depending on sulfur availability.
Serpentinisation of ultramafic rocks and the olivine occurs
90
PROCEEDINGS
Fig.7 Cores of altered tectonic breccia

Fig. 9 Scatter plot of total Ni versus S content, with bubbles to
sulfide nickel proportion (BrMeOH)
METALLURGICAL TESTWORK
Further studies were performed to figure out the
geometallurgical properties of Latao mineralized rocks.
First metallurgical testwork was conducted by PT
Sucofindo in Cibitung, West Java within two stages, i.e,
the stage 1 bulk flotation test work and stage 2
optimization testwork. In general, the stage 1 bulk
flotation test consists of standard rougher and cleaner
flotation at P80 105m (Figure 10, 11, 12). While, the
stage 2 optimization testwork consists of optimization
based on result from stage 1. The goal of optimization test
is to describe the recovery parameter of reagent testing
and relationship of grind versus recovery, grade versus
recovery, and time versus recovery kinetic test
(Sucofindo, 2012).
Fig. 8 Magnetite rock body outcrop
BRMEOH PARTIAL EXTRACTION ASSAY

Klock et al (1986) explained that when nickel sulfide
deposit are so low in grade as to be economically
marginal, they pose problem in ore grade evaluation
because the silicate bound is not recoverable. Klock et al
(1986) compared six analytical techniques for selective
chemical dissolution of sulfide with purpose of defining
the best method for accurately determining the
concentration of sulfide bound nickel. As result, the
bromine-methanol dissolution (BrMeOH) method was
recommended for assaying the nickel sulfide deposit when
nickel is also present in silicate phases. Therefore, the
BrMeOH was selected because of its capability to separate
the recoverable sulfide nickel with the non-recoverable
oxide and silicate nickel.
Generally, the average nickel grade of Latao is about
0.2%, with sulfide nickel amount obtained from brominemethanol dissolution at about 0.1%. According to the
BrMeOH assay, the sulfide nickel amount in samples of
Latao vary up to 0.25%. The depths of the mineralized
rocks recorded from the drillholes vary as well from 4
221 m. Lithologically, the mineralized zones are mainly
hosted by the highly-serpentinized ultramafics and the
altered tectonic breccia. From geochemistry plot (Figure
9), results of BrMeOH partial extraction strongly correlate
with sulfur content. Where the sulfide nickel amount
based on BrMeOH assay is increasing, the sulfur contents
are increasing as well.
Fig.10 Generic work flow of the flotation test
In the stage 1 testwork, elemental assay performed on

feed resulted in Ni 0.40%, Fe 13.2%, S 0.05% and Mg
15.34%. Samples were then milled to P80 105m and four
types of reagent were used for the flotation test, i.e,
sodium isobutyl xanthate, methylisobutil carbinol, sodium
isopropropyl xanthate and methyl glycol ether. Flotation
test revealed that the highest Ni grade in the concentrate is
only 0.38% at 39.20% recovery (Figure 13), or in other
words, there are no upgrading during the metallurgical
testworks (Sucofino, 2012).
Microscopy locking and liberation analysis suggest the
present unliberated nickel sulfide minerals at size below
53m in the tailing, but majority of coarse nickel sulfide
mineral > 90m can be recovered using flotation process.
91
PROCEEDINGS
The result of stage 1 testwork was then become the

base for stage 2 - optimization testwork. It was performed
by lowering the grinding size to P80 45m and change
several parameter such as solid density and reagent
consumption. Compared to standard test, the optimization
was able to improve the recovery to 63.7% (Figure 15),
however there is essentially still no increase in the
concentrate grade, 0.4% Ni in feed vs 0.39 Ni in
concentrate (Sucofindo, 2012).
Second metallurgical testwork was carried out by SGS
laboratory at Lakefield, Canada. Two composites of
samples were submitted with feed nickel (total) grades are
0.21 and 0.22 and sulfide nickel sulfide content based on
BrMeOH are 0.091% and 0.092%. Testworks were
performed at P80 134 m and 70 m. The Ni assay of the
concentrate is same with the feed at 0.22%, or no
upgrading.
Fig.13 Cumulative Ni recovery versus cumulative MgO for

Latao samples (Sucofindo, 2012)
Fig.14 Microscopy analysis to tail samples shows 45m

heazlewoodite and 5m violarite locked in 160m serpentine
matrix (Sucofindo, 2012)
Fig. 11 : Flotation process (www.911metallurgist.com)
MINERAL LIBERATION AND LOCKING

CHARACTERISTICS
Mineral liberation and locking analysis was then
performed to investigate the problem why there is no
upgrading in the concentrate. For that purpose, Sucofindo
first carried out a microscopic analysis in 2012. They
suggested that the majority liberation size of the Nisulfides (millerites and heazlewoodite) was less than
50m. The ore also contain high serpentine matrix which
is highly altered to clay-size materials; very fine-grained
and fissile (breaks up easily). It means that finer grinding
will liberate the Ni-sulfide further, but it will also increase
the number of fines that will make thick slurry. The high
content of clay size (<10m) material in the slurry hinders
reagent absorption on the surface of Ni-sulfide minerals.
Mineral examination on feed samples also suggested
highly nickel oxidized serpentine rock. The trace amounts
of nickel sulfide mineral that are present are secondary
alteration products from primary nickel sulfide
(pentlandite) of which no longer exist in its original form
of chemical composition. This highly oxidized ore does
not respond well to sulfide flotation using xanthate
(Sucofindo, 2012).
Fig.12 Flotation test of Lataos samples at Sucofindo

Laboratory 2012
The results of these metallurgical testworks were offcourse far below the company expectation. If the
Ronnbackens samples were reported able to produce
concentrate with a grade of 28% at 80% recovery, no
nickel upgrading was resulted from Lataos samples in the
metallurgical testwork.
92
PROCEEDINGS
storaging. Vast oxidation may have occurred as well

during preparation when samples cut or crushed to minus
10 Tyler mesh at PTVI sample preparation facility. After
the crushing, these samples were usually stored for about
2 3 weeks before shipped to laboratory for metallurgical
testwork. The fine-grained crushed material during storage
time may allow access for oxygen to oxidize the sulfide
minerals. Finally, it may reduce capability of sulfide
mineral liberation during the flotation test.
OPPORTUNITIES
Latao is an example of new frontier for nickel
exploration in Indonesia. In addition to Latao, PT Vale
Indonesia Tbk also found trace of nickel sulfide
occurrence in Sorowako, South Sulawesi. In Sorowako,
nickel sulfide minerals are also occurring disseminated in
serpentinites (Figure 16). Millerites and pentlandites are
observed megascopically and larger in size than that in
Latao. Meanwhile, violarites and heazlewoodites are also
observed microscopically. However, the nickel sulfides in
Sorowako are occurring in different geological setting
than Latao. They occur in the serpentinite block within an
olisostrome complex of a mlange complex. Therefore,
finding a massive ore body will be difficult since the
serpentinite blocks are highly scattered. These two
findings show that it is likely possible to find another type
of nickel sulfide deposits in Indonesia. Nevertheless, the
nature of the nickel sulfide minerals shall be wellunderstood prior to exploring further the nickel sulfide
deposit. Thus, a correct method can be applied during
exploration. Hopefully, lesson learnt from Latao can be a
useful information when dealing with nickel sulfide
deposit.
Fig. 15 Cumulative recovery obtained from flotation test at feed

P80 45 m grinding size, compared to the standard P80 150 m
grinding (Sucofindo, 2012)
Furthermore, high definition mineral liberation and

locking characteristics study using XRD, QEMSCAN and
EMPA (electron microprobe analyzer) was conducted at
SGS laboratory in Canada. The study concluded that the
pentlandite show poor liberation. Combination of free and
liberated pentlandite accounts for ~14% of the
pentlandites in the sample. Roughly 41% of the nonliberated pentlandite is found as inclusion in serpentine;
24% appears in complex particles of two or more other
minerals; and 15% are in ternary particles with olivine and
serpentine. The maximum pentlandite liberation was 40%
at ~15m. above 15m in size, liberation will drop
significantly due to the nature of the pentlandite inclusion
in serpentine (SGS Canada, 2012).
LESSON LEARNT FROM LATAO
So far, primary magmatic nickel sulfide deposit is still
the most interesting target for nickel sulfide exploration.
When dealing with other type deposit than the primary
magmatic, such as the Latao, Ronnbacken and Dumont,
careful study must be undertaken to better understand the
nature of the nickel sulfide minerals. Detail geological
structures position, orientation and dipping in the study
area should also be studied to better predict the possible
channeling behaviors of the hydrothermal fluids and
geometry of the ore bodies.
Actually, the grain size of nickel sulfide minerals of
Latao does not differ much from the Ronnbackens.
Millerites, pentlandites and heazlewoodites of Latao vary
in sizes, but mostly are ranging between 10 20 m.
Meanwhile Ronnbacken project reported that the majority
of nickel-bearing sulphides gain size fall within range of
15 to 50 m with average closer to 25 m (SRK
Consulting, 2011). Considering to this fact, two possibility
emerge as the root cause why the flotation test fail to
achieve high nickel grade concentrate. First possible
answer is because the nature of the nickel sulfide minerals
that are present as inclusions in serpentine, or locked by
the serpentine. The second possible answer is because of
the sulfide minerals in the feed samples had been
oxidized. The second possible root cause is probably
related with the sample handling and sample preparation
method. Nickel sulfide minerals like millerites and
pentlandites are easily to oxidize.
During the exploration, Lataos core samples were
wrapped with plastics, but possibly the wrapping method
is still inadequate to prevent oxidation during the
Fig. 15 : Disseminated millerite needles in Sorowakos

serpentinite
CONCLUSION
Finally, lesson learnt can be drawn from Latao nickel
sulfide exploration are:
1) So far, primary magmatic nickel sulfide deposit is
still the most interesting target for nickel sulfide
exploration. When dealing with other type of deposit
than the primary magmatic, detail structural geology
study must be carried out to understand the
channeling behavior of the hydrothermal fluids and
predict the ore body geometry.
2) Aside to nickel grade, nature of nickel sulfide
mineralogy shall also be the primary consideration
when exploring a nickel sulfide deposit, particularly,
93
PROCEEDINGS
3)
4)
5)
Hall, R., 1996. Tectonic Evolution of Southeast Asia, London

Special Publication, London
Hall, R., Wilson, M.E.E, 2000. Journal of Asian Earth Sciences
18, 781 808
Hinchberger, F., Malod, J.A, Rehault, J.P, Villeneuve, M.,
Royer, J.Y, Burhanuddin, S., 2005. Tectonophysics 404, 91
- 118
Kadarusman, A., 2008. internal presentation. PT INCO, Tbk
Kadarusman, A., Miyashita, S., Maruyama, S., Parkinson, C.,
Ishikawa, A., 2004. Tectonophysics 392, 55 83
Klock, P.R., Czamanske, G.K., Foose, M., Pesek, J., 1986.
Chemical Geology 54, 91 - 118
Lightfoot, P., Vandenburg, B., 2013. internal report. Vale Base
Metal
Rafianto, R., Attong, F., Matano, A., Noor, M.E.S., 2012.
Majalah Geologi Indonesia, Vol. 27, 87 107
Rusmana, E., Sukido, Sukarna, D., Haryanto, E., Simandjuntak,
T.O., 1993, Pusat Penelitian dan Pengembangan Geologi,
Bandung
SRK Consulting, 2011. report prepared for Nickel Mountain
Resources AB, Sweden
Sucofindo, 2012. report for PT Vale Indonesia, Tbk,
SGS Canada, 2012. report for PT Vale Indonesia, Tbk, Indonesia
of which related with the mineral liberation and

locking characteristics
Size of nickel sulfide minerals is a critical
metallurgical parameter. The smaller in size, the more
difficult to liberate in a flotation.
Sample handling and sample preparation method
must adequate to prevent the oxidation of the nickel
sulfide minerals.
Nickel sulfide mineralization in Latao and Sorowako
show that it is likely possible to find another type of
nickel sulfide deposit in Indonesia. However, the
nature of the nickel sulfide mineral should be wellunderstood prior to exploring further the nickel
sulfide deposit.
REFERENCES
Foose, M.P., Zientek, M.L., Klein, 1995. USGS, Denver,

Colorado
Gahlan, A.H., Shoji, A., Ahmed, A.H., Ishida, Y., Abdel-Azis,
Y.M., Rahimi, A., Journal of African Earth Sciences 46,
318 - 330
94
PROCEEDINGS
95
PROCEEDINGS
Beneficiation and Mineralogical Study of Bauxite Deposits in Ketapang, West

Kalimantanfor Optimum Bayer Alumina Refinery Process
Robby I. Rafianto1, Henry A. Cahyono2, Abimanyu Yudhaswita2, Alan Matano3, Yusni Marta2
1
PT Cita Mineral Investindo Tbk.

PTHarita Prima Abadi Mineral
3
PT Harita Exploration
ABSTRACT
Bauxite is the principal raw material for the production of alumina and aluminium metal. The exploration and mining
activities for Bauxite deposits are known in Indonesia since pre-World War II. A smelter grade alumina (SGA) refinery is
built in Ketapang Regency, West Kalimantan by PT. Well Harvest Winning (WHW) as a purpose of increasing the valueadded to bauxite mineral through domestic Bayer alumina processing. The specification of certain bauxite quality is
required by WHW and a beneficiation and mineralogical study is required to meet metallurgical criteria that support
economic Bayer refinery processing.
The bauxite deposits in Ketapang are a lateritic blanket of weathering product from Early to Late Cretaceous of granite
monzonite-diorite plutonic and volcanic series which are rich in alumino-silicate minerals. The combination of PSD, XRF,
wet chemistry and semi quantitative XRD mineralogical data is effective to assess grain sizes and mineralogical behaviors
during screening and washing process. At screen size +1 mm boundary the values of total Al2O3, available Al2O3, LOI and
Tal2O3/TSiO2 ratio are significantly increased while TSiO2, RSiO2 and Fe2O3 values are decreased. The identified minerals
by XRD in each fraction size are gibbsite, goethite, magnetite, hematite, anatase, quartz, kaolinite, illite and paragonite.
The principal bauxite mineral found is only gibbsite while boehmite and diaspore are not identified. The 2014 and 2016
beneficiations studies shown that the washing and screening processes are still not optimum in separating fine grained
gibbsite and interlocked kaolinite minerals. For long-term economic benefits, mapping of bauxite ore types are quite
important, then the beneficiation plant at mine fronts can be optimized and impacting for low cost Bayer processing.
the washing plant to produce metallurgical grade bauxite
(MGB). A certain number bauxite deposits have been
found to meet the stringent physical and chemical
specifications expected of smelter grade alumina in CITA
Air Upas and Sandai mine concessions. A preliminary
beneficiation study was made in 2014 (LAPI ITB) with
samples taken from mine faces, mine washed bauxites and
tailings. The 2016 study is covering more representative
areas based on drillhole samples. The specification of
certain bauxite quality is required by WHW and a
beneficiation and mineralogical study is required to meet
metallurgical criteria that can support economic Bayer
refinery processing.
This paper is a compilation and more focus on
beneficiation and mineralogical assessment in Air Upas Sandai areas by CITA geologists to investigate bauxite
specifications required for Bayer process under PT. WHW
operation.
INTRODUCTION
Bauxite was first discovered by French geologist
Pierre Berthier in 1821, near the village Les Baux-deProvences in the French Alpilles (Gendron et al, 2013). It
is a principal raw material for the production of alumina
and aluminium metal. In Indonesia, the initial discovery of
bauxite in Southeast Bintan was made in 1924
(Bemmelen, 1940 in Rodenburg, 1984). Bauxite
production in Indonesia was started in 1935 by the
Netherlandsch Indische Bauxite Exploitatie Maatschappij
(Rodenburg, 1984) and continue by several companies
until now. The major bauxite deposits in Indonesia are
reported from Riau Archipelago and West Kalimantan.
The geology and genesis of Indonesian bauxite deposits
were described by Bemmelen (1949), Rodenburg (1984),
Surata et al (2010) and Toreno et al (2012).
From an industrial perspective, bauxite is a raw
material that can be economically processed to high purity
alumina (>98% Al2O3) in the Bayer process. A mandatory
domestic minerals processing is required under Law No. 4
of 2009 in Indonesia. Under this obligation, a smelter
grade alumina (SGA) refinery is built in Ketapang
Regency, West Kalimantan by PT. Well Harvest Winning
(PT. WHW) as a purpose of increasing the value-added to
bauxite mineral through domestic processing. PT. WHW
requires consistent bauxite supply from several mines in
Ketapang under agreement with PT. Cita Mineral
Investindo Tbk. (CITA) as a bauxite miner. Currently
CITA is mining and beneficiates the bauxite ore through
THE BAYER PROCESS

Production of aluminium metal from bauxite is a two
stage process. The first stage involves the refining of
bauxite ore through the Bayer process and resulting in the
production of alumina. The second stage involves the
electric reduction of alumina to aluminium metal through
the Hall-Heroult Process. This paper is focus only on the
alumina production through the Bayer Process. The Bayer
process is the principal method for production of alumina
from bauxite worldwide.
96
PROCEEDINGS
The Bayer process, patented in Germany by Karl

Joseph Bayer in 1888, exploits the relatively high
solubility of aluminium oxide minerals in hot caustic soda
solution. Separation of the insoluble phases, followed by
gibbsite precipitation and calcination of the gibbsite to
alumina (Smith, 2008). Operations with a red background
represent those involving either the bauxite or mud, and
thus represent the so-called red side of the Bayer
Process (Fig. 1). Operations with white background relate
to processes in the absence of these solid materials, and
represent the so-called white side operations
(Smith,2008).
caustic soda such as kaolinite clays and fine-grained

quartz, cause caustic soda losses in the Bayer Process.
Two to three tonnes of bauxite, depending on its
composition, are required to produce one tonne of alumina
(Fig. 2). The cost of bauxite is a multiple of the Bauxite
Consumption Factor and the unit cost of bauxite, and is a
major cost item at most alumina plants. Bauxite
consumption factor can be reduced by increasing alumina
recovery during processing (Chin, 1984). The typical
major cost items at most alumina plants is shown in Fig. 3.
The definition of terms used in bauxite and alumina
industry listed in Table 2
Fig. 2. Process flow and the typical recovery of material during

processing from bauxite to aluminium.
Fig. 1: Simplified schematic of the basic Bayer Process

(Modified from Smith, 2008).
The actual processing conditions, such as the leach

temperature, digestion and soda caustic consumption are
greatly influenced by the type of bauxite to be processed
(Hill and Sehnke, 2006). The ideal bauxite characteristics
(MGB) for Bayer plant feed are listed in Table 1.
Certain features of bauxite deposits control the
efficacy of the Bayer Process. The most important are the
relative amounts of the alumina-bearing minerals and the
presence of deleterious minerals which also react with
caustic soda (Gow and Lozej, 1993). The principal
aluminium hydroxide minerals found in varying
proportions within bauxite are gibbsite Al(OH)3 and the
polymorphs boehmite and diaspore, both ALO(OH) (Hill
and Sehnke, 2006). The trihydrate gibbsite is most easily
soluble in caustic soda requiring a digestion temperature
of no more than 140o C. On the other hand, the
satisfactory solubility of boehmite is normally obtained
above 200o C, while diaspore is even more intractable,
requiring temperatures of the order of 300o C (Andrews,
1984). Other minerals in the ore which react with the
Fig. 3. Typical major structure cost items at most alumina plants.
GEOLOGICAL SETTINGS
Regional Geology
The island of Borneo (or Kalimantan of Indonesian
part) presently lies upon the southeastern margin of the
greater Eurasian plate. It is bounded to the north by the
South China Sea marginal oceanic basin, to the east by the
Philippine Mobile Belt and the Philippine Sea Plate and to
the south by the Banda and Sunda arc systems. It is
bounded to the west by the Sunda Shelf and ultimately by
Paleozoic and Mesozoic continental crust of the Malay
Peninsula. The Greater Kalimantan Block is surrounded to
the north, east, and south by plate boundaries and arc
systems which are presently active or which have been
active during the Tertiary and it is bounded to the west by
an underexplored shelf region which possibly conceals a
terrane boundary (Darman & Sidi, 2000).
Kalimantan can be divided into several roughly E-W
trending tectonic provinces (Fig. 4). The northern portion
97
PROCEEDINGS
of the island is dominated by the Cretaceous and Eocene

to Miocene Crocker-Rajang-Embaluh accretionary
complex. This consists primarily of turbidites which were
being shed northeastward (present day coordinates) off of
the Schwaner and younger volcanic arcs into a paralic to
deep marine trench basin. These sediments were
imbricated, deformed, and weakly metamorphosed during
Cretaceous and Tertiary subduction and finally were
intruded by late stage and post subduction intrusions of
the Oligo-Miocene Sintang Group.
A core of Paleozoic or older continental crust
metamorphic rocks in the SW part of the island. The
metamorphic rocks are intruded by biotite granite which
yields K-Ar ages from Permian to Late Triassic. Many of
the granitic rocks contain a strong foliation, and the Late
Triassic ages are obtained from biotites from deformed
rock The Permian dates come from hornblende crystals
from undeformed regions of the granites or from
amphibolite en-claves. The older ages are interpreted as
minimum intrusive ages and the Middle to Late Triassic
ages as the deformation age of the suite (Darman & Sidi,
2000).
According to geological maps of Kendawangan
(Sudana et al., 1994), Ketapang (Rustandi & Keyser,
1993) and Pontianak/Nangataman (Sanyoto and Pieters,

1993) sheets made by Geological Research and
Development Centre of Indonesia, the concession area has
a several formations consist of Laur Granite, Sukadana
Granite, Kempari Sandstone, Kerabai Volcanics, Sintang
Intrusives and Alluvial Deposits which classified as
follows (Fig. 5):
1) Laur
Granite
(Kll):
hornblendebiotite
monzogranite, rare biotite syenogranite and biotite
hornblende granodiorite
2) Sukadana
Granite
(Kus):
monzogranite,
syenogranite, quartz monzonite, quartz syenite, alkali
feldsfar granite, and rare granodiorite, tonalite, quartz
diorite. Rocks contain variable amounts of biotite and
hornblende and occasionally clinopyroxene and alkali
amphibole
3) Kempari Sandstone (Kuke): quartz arenite and
lithic arenite
4) Kerabai Volcanics (Kuk): andesite, dacite and basalt
lava, lava breccia, pyroclastic and small intrusives
5) Sintang Intrusives (Toms): porphyritic andesite
6) Alluvial (Qa): mud, sand, gravel and plant matters
Table 1. Ideal characteristics for metallurgical-grade bauxite (Bayer plant feed)
Table 2: Definitions of terms used in this study
Ideal Grade Characteristic

Low "reactive silica" (>1.5% - < 3.0%) (kaolinite)
Alumina
High extractable alumina (>49%)

Low organic carbon
Low boehmite (<3%)
Low goethite (tolerable in a high-temperature
plant or with high hematite)
Impact if Inadequate
Increases casutic usage, a critical operating cost factor
Increases capital and operating costs for mining, processing, and especially mud disposal (larger
equipment, bigger containment dams, and increased materials handling
Increases operating costs by reducing plant efficiency and lowers product quality. Alternatively
increases capital costs by adding organics removal facility
Precludes low-temperature processing that can increase capital and operating costs
Slows clarification, lowers product quality, and increases alumina loss via mud circuit. Increases
capital (equipment size) and operating costs (plant efficiency)
Aluminium
Available Alumina (AA)
Bauxite
Low moisture (can create dust nuisance if too low) Increases capital costs (larger evaporation facility), fuel consumption, and shipping costs
Iron content (ideally > 5% - < 15%)
Low iron can lower product quality. High iron dilutes alumina content of bauxite
Low quartz
Increases maintenance costs (pipe wear). Increases caustic usage in high-temperature plants
Can lower process efficiency (sulfur, chlorine, calcium) and metal quality (gallium, zinc,
vanadium, phosphorus
Low impurities and trace elements
Soft and friable
Increases mining and grinding costs
Increases capital (larger digestion equipment) and operating costs (reduces plant efficiency,
longer residence times)
Dissolves readily
Low titania
Can increase caustic usage in high-temperature plants
Low carbonates
Can require special processing
Bauxite Ore (Metallurgical Grade

Bauxite)
Reactive Silica (RSiO2)
High-purity aluminium oxide (Al2O3). Pure alumina (100%Al2O3) contains 52.9%

aluminium and 47.1% oxygen
Aluminium metal is the thirteenth element of the Periodic Table. It is produced by
electrolytic smelting of alumina in the Hall Heroult plant
The amount of alumina extractable in solution from bauxite by the Bayer process, a hot,
high-pressure, caustic soda (NaOH) leach
Bauxite is a soil or rock formation which is composed mainly of aluminium hydroxide
minerals. From an industrial perspective, bauxite is a natural material from which alumina
can be extracted in a Bayer plant
A bauxite body which is economically mineable at present or in the foreseeable future.
Currently, its composition would normally be >45% Al2O3, <20% Fe2O3 and <5%SiO2
Silica phases present in the bauxite, generally as clays and fine-grained quartz, which
cause caustic soda loss in the Bayer Process. Quartz, if present as coarse silt or sand, is
unreactive in the Bayer Process
(Modified from Gow and Lozej, 1993)
(Hill and Sehnke, 2006)
98
PROCEEDINGS
silicate minerals. Mineralised areas are prominent along

the marshy flanks of these igneous suites at the west and
south east of the island. Based on petrographic analysis,
the bauxite laterites host rocks in CITA mine concessions
are consist of quartz-monzonite, quartz-monzodiorite,
biotite monzonite, granodiorite, diorite and breccia with
intermediate composition (NickelPhil, 2012).
Bauxites mainly occur in tropical zones that are
subjected to intense weathering conditions where
oxidations and water saturation is high. Such conditions
enable the dissolution of the kaolinite and the precipitation
of hydrous alumina silicate in the weathering process.
One of the main factors in forming good bauxite
laterite is geomorphology aspects. The slope of the
landform plays an important part in bauxite laterite
formation intensity. Flatter slope will facilitate the surface
water to infiltrate easier, thus makes the weathering
process work intensively. Weathering process is the main
process to form a bauxite laterite enrichment. On the
contrary, the high slope will facilitate the run-off water,
means less water infiltrate the soil or the rock. In turn,
does not facilitate the weathering process to form the
bauxite laterite.
The triangular variation diagrams of Al2O3-SiO2-Fe2O3
are showing the degree of lateritization, mineral control
and bauxite classifications. Fig. 6 from Air Upas area
indicates that Al2O3-rich composition is indicative of a
higher degree of lateritization while SiO2-rich samples
experienced weak lateritization (Meyer et al. 2002 in Gu
et al., 2013). Based on mineralogical classification by
Aleva, (1994 in Gu et al., 2013). The moderate to strong
lateritization samples in Figure 6 have correlations with
classification of high grade bauxite bauxite kaolinitic
bauxite and ferritic bauxite (Fig. 7) and their distributions
in Air Upas are shown in Fig. 8.
Fig.5. Regional geology of Kendawangan, Ketapang and

Pontianak with CITA mine concessions (Modified from Sudana
et al., 1994, Rustandi et al., 1993 and Sanyoto at al., 1993).
During the Early Permian - Late Triassic, the oldest

rock in the region, metamorphic rocks were formed from
regional metamorphism in shallow marine sedimentary
rocks with varying degrees of metamorphic. At the
Jurassic - Cretaceous, sedimentation process occurs within
the shallow sea near the coast and paralic environment.
At the Late Jurassic, two intrusions were occurred
namely Belaban granite and Sukadana granite that cut the
metamorphic rocks. Furthermore during the Cretaceous
Paleocene, there were 2 magmatism events have been
taken place in the Early Cretaceous, as follows: Laur and
Sukadana granites, and Sangiang granite. Kerabai
volcanic rocks are co-magmatic with those granites,
Bunga basalt as well as mafic and felsic dikes are
widespread cut the granites.
The whole process magmatism reflects the transition
between tectonic compressions with stretching process. In
the process there is an intrusion granite tectonic
compression associated with subduction in the Lower
Cretaceous, is the stretching process is still ongoing
process late magmatism, where the rocks appear intrusion
Sintang and volcanic plugs during the Oligocene Miocene. The occurrence of intrusion Sukadana Granite
conjunction with the appointment process, where the
sedimentary rocks also come up so that in some
sedimentary rocks are folded with a slope of between 300
to 700.
Fig. 6. Triangular diagram of bauxite lateritization degree in Air

Upas area (Modified from Meyer et al. 2002 in Gu et al., 2013).
Local geology of bauxite occurences

The bauxite deposits in all of the prospects are lateritic
blanket of weathering products from granite-monzonitediorite series of plutonic petrogenesis in West Kalimantan.
These rocks are acidic igneous rocks that rich in alumino99
PROCEEDINGS
TEST WORK RESULTS

Particle Size Distribution Analyses
Bauxite drillhole samples were collected in each
meter. Every samples was weight in wet stage and dry
stage at 105C to get moisture content (MC). Samples
within one hole were composited and homogenized to get
significant weight for Particle Size Distribution (PSD)
testwork. Each composited sample is about 10 15 kg of
dry sample from representative bauxite horizon of each
drill hole. The PSD testwork conducted in wet stage, and
reducing from initial of 18 screens to 8 screens for
effectiveness of the testwork. The screen sizes are various
from 50 mm to 0.3 mm. Based on PSD analyses, it is very
clear that the bauxite size from Air Upas and Sandai areas
are divided into two major populations with boundary of
1mm screen size. The individual fraction retained weight
(%) analyses are supporting this evidence. If fine grained
removed with boundary of 1 mm, the retained weight (%)
recovery is about 60% (Fig. 9 and Fig. 10). This figure
can be identical with the Concretion Factor (CF) of
washed bauxite.
Fig. 7. Triangular diagram of bauxite mineralogical classification

in Air Upas area (Modified from from Aleva, (1994 in Gu et al.,
2013).
Sampling and analytical methods

For this mineralogical study, a test work package is
conducted on bauxite resources from 31 twinhole drilling
samples accros the project areas of Air Upas and Sandai.
The test work includes:
a. Particle Size Distribution (PSD) at 8 screen sizes at
the wet stage and dry weight. The wet stage is
relatively equal with bauxite washing process.
b. Assaying in each fraction using: XRF for TAl2O3,
TSiO2, Fe2O3, TiO Wet Chemistry + Bomb digest
calorimeter for RSiO2, Av Al2O3
- Gravimetry for
LOI
c. Mineralogy Study at composite fraction size by semi
quantitative XRD analyses
In 2014 CITA made a preliminary beneficiation study
(LAPI ITB, 2014), the samples were taken from unwashed
mine front, washed bauxite product and tailing. All data
from 2014 and 2016 studies were compiled and analyzed
by Kadarusman (2016) as CITA principal consultant for
Bauxite exploration.
Fig.9. PSD distributions from Air Upas (top) and Sandai

(bottom) drillhole samples. There are two major populations:
fine-grained 1mm size and coarse-grained +1mm size.
Fig. 8. Bauxite type map based on mineralogical classification

in Air Upas area.
100
PROCEEDINGS
Fig. 11. Average chemistries vs individual screen size of Air

Upas area
Figure 12. Average chemistries vs individual screen size of

Sandai area
The estimated % of available alumina (AA) from total

alumina (Tal2O3) are estimated based on wet chemistry
and stoichiometric calculation. For this exercise the
washed and unwashed samples were compared. The AA
value is quite important information for Bayer process as
it is indicates the real amount of alumina can be extract in
caustic soda solution. As expected, the washed samples
relatively have higher AA values than unwashed samples.
It is shown that Air Upas area is relatively higher in AA
than Sandai area (Fig. 13). This is assumed that the
bedrock of laterite profile has controlling the alumina
content during lateritization process. Unfortunately,
almost of the previous test pit and drillholes were not
deeply enough to sampling the bedrock. The further study
is required to investigate this hypothesis.
Figure 10. Average recovery vs each screen size for Air Upas
(top) and Sandai (bottom).
XRF and Wet Chemistry

This testwork is combining XRF, wet chemistry and
PSD data. Each fraction of PSD data was analyzed by
XRF and wet chemistry for Total TAl2O3, available Al2O3,
Total TSiO2, Reactive RSiO2, Fe2O3, MgO, P2O5, TiO2
and Loss on Ignition (LOI).
The combined data for average Air Upas and Sandai
are shown in Fig. 11 and Fig. 12. From both Air Upas and
Sandai Figues, there are connections between PSD, XRF
and wet chemistry. At screen size +1 mm boundary the
values of Total Al2O3 (Tal2O3), available Al2O3 (AA),
LOI and Tal2O3/TSiO2 ratio are significantly increased
while TSiO2, RSiO2 and Fe2O3 values are decreased. This
data evidence is a very important information for the
purpose of bauxite beneficiation plant design prior feed to
the Bayer refinery plant.
Fig. 13. Available Alumina(AA) composition against Total

Alumina(TA) from washed and unwashed samples.
Semiquantitative XRD Mineralogy Analyses

Four representative drillhole samples from Air Upas
and Sandai are selected for Semiquantitative XRD
analyses. The objective is to assess the mineralogy
distribution of each fraction from -0.3 mm to +50 mm of
composited samples.
The identified minerals in each fraction size are
gibbsite, goethite, magnetite, hematite, anatase, quartz,
kaolinite and paragonite. The principal bauxite mineral
found is only gibbsite while boehmite and diaspore are not
identified. This data is encouraging as gibbsite is the most
expected bauxite mineral because it is easily soluble in
101
PROCEEDINGS
Bayer Process (lower temperature process).

The
abundance of kaolinite occurrences are not expected due
to identical with reactive silica that can increase caustic
soda usage and higher processing cost.
Comparing with the existing washing plant in CITA
mine operations, It is normally expected that after washing
and screening process, gibbsite is concentrated in coarse
fraction of + 2mm while kaolinite is concentrated in finer
fraction of -2 mm. However, surprisingly the data shows
that fine-grained gibbsite minerals are still exists in finer
2 mm fractions while kaolinite minerals are also exists in
+2 mm coarse fractions (Fig. 14, 15, 16 and 17). It is
possible that the existing screening washing process
both in laboratory and mines are not optimum to separate
fine-grained gibbsite and kaolinite.
Fig. 15. Air Upas: fine-grained gibbsite minerals are still exists
in finer 2 mm fractions after washing screening while
kaolinite minerals are also exists in +2 mm coarse fractions.
Fig. 14. XRD mineralogy and chemistries compositions for each

fraction size from Air Upas area
Fig.16. XRD Mineralogy and chemistries compositions for each

fraction size from Sandai area.
102
PROCEEDINGS
Fig. 18. Washing plant performances and major chemistries at

mine sites of Sandai, Labai and Air Upas. Front = insitu mine
bauxite samples, Olahan = washed bauxite samples (LAPI ITB,
2014).
For long-term economic benefits, mapping of bauxite

ore types are quite important. It is interpreted that certain
bauxite type is naturally clean without kaolinite and the
other type is interlocked with kaolinite. By support from
Geology Department, the mine plan can be optimized to
support crush no crush and wash no wash
decisions during mine operations. Then, the bauxite mine
tonnage can be optimized and minimized fine-grained
bauxite loss to the tailing dams.
CONCLUSION
Figure 17. Sandai: fine-grained gibbsite minerals are still exist in

finer 2 mm fractions after washing screening while kaolinite
minerals are also exists in +2 mm coarse fractions.
The major conclusions from this study are summarized

below.
The moderate to strong bauxite lateritizations in
Ketapang have correlations with the classification of
high grade bauxite bauxite kaolinitic bauxite and
ferritic bauxite.
Based on PSD analyses, it is very clear that the
bauxite size from Air Upas and Sandai areas are
divided into two major populations with boundary of
1mm screen size. If fine grained removed with
boundary of 1 mm, the retained weight (%) recovery is
about 60%.
At screen size +1 mm boundary the values of total
Al2O3, available Al2O3, LOI and Tal2O3/tsio2 ratio
are significantly increased while tsio2, rsio2 and Fe2O3
values are decreased. This data evidence is a very
important information for the purpose of bauxite
beneficiation plant design prior feed to the Bayer
refinery plant.
As expected, the washed samples relatively have
higher AA values than unwashed samples. It is shown
that Air Upas area is relatively higher in AA than
Sandai area. This is assumed that the bedrock of laterite
profile has controlling the alumina content during
lateritization process.
The identified minerals by XRD in each fraction size
are gibbsite, goethite, magnetite, hematite, anatase,
quartz, kaolinite, illite and paragonite. The principal
bauxite mineral found is only gibbsite while boehmite
and diaspore are not identified. This data is
encouraging as gibbsite is the most expected bauxite
mineral because it is easily soluble in Bayer Process
(lower temperature process).
Surprisingly the data shows that fine-grained
gibbsite minerals are still exists in fine fractions of 2
mm while kaolinite and illite minerals are also exists in
DISCUSSION
The main objectives of this beneficiation study are
assessing the optimum methods to reduce silica, increase
alumina content effectively and get optimum tonnage
recovery. Additionally, the study results can support
washing plant improvements at mine fronts. The bauxite
beneficiation is a quite challenging process as the types of
bauxites are various, depending on bedrock chemistries
and lateritization processes.
The combination of PSD, XRF, wet chemistry and semi
quantitative XRD mineralogical data is effective to assess
grain sizes and mineralogical behaviors during screening
and washing process. The data from 2014 and 2016
studies show that the existing beneficiation methods both
at lab scale and washing plant are not optimum to recover
fine-grained gibbsite and potentially loss to tailing (Fig.
18). On the other side, some kaolinites are also locked in
coarse-grained bauxite and potentially include in bauxite
feed for Bayer process. The kaolinite clays are identical
with reactive silica and can cause caustic soda losses in
the Bayer Process, this situation can impact to higher
processing cost. This study is strongly support for
beneficiation improvement at mine sites with main
objectives to increase alumina content and lowering
deleterious elements such as silica. For further study,
crushing, scrubbing and washing methods are need to be
considered to remove kaolinite with more efficient.
103
PROCEEDINGS
Hill, V. G., Sehnke, E., 2006, Bauxite, in Industrial Minerals and

Rocks: Commodities, Markets and Uses, Edited by J.
Kogel, N. Trivedi, J. Barker, S. Krokowski, Littleton, CO:
Society for Mining, Metallurgy and Exploration, p. 227
261.
Kadarusman, A., 2016, Particle Size Distribution and
Beneficiation Study for Bauxite, Internal Presentation for
PT. Cita Mineral Investindo Tbk. 47 p.
LAPI ITB, 2014, Laporan Akhir: Kajian Analisis Benefisiasi
Bauksit, Internal report for PT. Labai Tambang Pertiwi, 38
p.
Nickel Phil, 2012, Mineral Resource Report Based on JORC
Code, North, Central and South-2 Bauxite Regions, West
Kalimantan Indonesia for PT. Cita Mineral Investindo Tbk.,
Part 1 of 2 Parts, 64 p.
Rodenburg, J. K., 1984, Geology, Genesis and Bauxite Reserves
of West Kalimantan, Indonesia, in Bauxite: Proceedings of
the 1984 Bauxite Symposium, Edited by L. Jacob Jr. New
York: SME AIME, p. 603 618.
Rustandi, E., De Keyser, F., 1993, Geological Map of the
Ketapang Sheet, Kalimantan scale 1 : 250,000. Geological
Research and Development Centre, Bandung.
Sanyoto, P and Pieters, P.E., 1993, Geological Map of the
Pontianak / Nangataman Sheet, Kalimantan scale 1 :
250,000. Geological Research and Development Centre,
Bandung.
Smith, P., 2008, High Silica Processing: Economic Processing of
High Silica Bauxites Existing and Potential Processes,
Parker Centre, CSIRO Light Metals Flagship, 42 p.
Sudana, D., Djamal, B., Sukido, 1994, Geological Map of the
Kendawangan Sheet, Kalimantan scale 1 : 250,000.
Geological Research and Development Centre, Bandung.
Surata, M., Suksiano, O., Pratomo, M., Supriyadi, 2010,
Discovery and Its Genetic Relationship of Bauxite Deposit
in Mempawah and Landak Regency West Kalimantan
Province, in Kalimantan Coal and Mineral Resources 2010
Proceedings, MGEI IAGI, p. 107 116.
Toreno, E. Y., Moetamar, 20112, Karakteristik Cebakan
Bauksit Laterit di Daerah Sepiluk-Senaning, Kabupaten
Sintang, Kalimantan Barat, in Bulletin Sumber Daya
Geologi Volume 7 No. 2, 2012, p. 1 - 22.
Van Bemmelen R. W., 1949, The Geology of Indonesia Volume
II, Economic Geology, Government Printing Office, The
Hague, 265 p.
+2 mm coarse fractions. It is possible that the existing

screening washing process both in laboratory and at
mine fronts are not optimum to separate fine-grained
gibbsite and kaolinite. For further study, crushing,
scrubbing and better washing methods are need to be
considered to remove kaolinite-ilite with more
efficient.
This study is strongly support for beneficiation
improvement at mine sites with main objectives to
increase alumina content and lowering deleterious
elements such as silica
For long-term economic benefits, mapping of
bauxite ore types are quite important. It is interpreted
that certain bauxite type is naturally clean without
kaolinite and the other type is interlocked with
kaolinite. Based on this bauxite type map, the
beneficiation plant at mine fronts can be optimized and
impacting for low cost Bayer processing.
REFERENCES
Andrews, W. H., 1984, Uses and Specifications of Bauxite, in
Bauxite: Proceedings of the 1984 Bauxite Symposium,
Edited by L. Jacob Jr. New York: SME AIME, p. 49 66.
Chin, L. A. D., 1984, Research Directions to Increase Alumina
Recovery from Bauxites, in Bauxite: Proceedings of the
1984 Bauxite Symposium, Edited by L. Jacob Jr. New
York: SME AIME, p. 641 650.
Darman, H. and Sidi, F.H., 2000, An Outline of the Geology of
Indonesia, Indonesian Association of Geologists (IAGI),
Jakarta, 192 p.
Gendron R. S., Ingulstad M., Storli E., 2013, Aluminum Ore:
The Political Economy of the Global Bauxite Industry,
Sample Material, UBC Press Vancouver Toronto, 24 p.
Gow, N. N., Lozej, G. P., 1993, Bauxite, Geoscience Canada
Volume 20 Number 1, p. 9 16.
Gu, J., Huang, Z., Fan, H., Jin, Z., Yan, Z., Zhang, J., 2013,
Mineralogy, geochemistry, and genesis of lateritic bauxite
deposits in the Wuchuan-Zhengan-Daozhen area, Northern
Guizhou Province, China, Journal of Geochemical
Exploration 130, p. 44 59.
104
PROCEEDINGS
105
PROCEEDINGS
Magmatic Fe-Ni-Cu Sulphides Occurence in Sebuku Island

Yoseph C. A. Swamidharma
PT Sebuku Iron Lateritic Ores
ABSTRACT
Magmatic mineralization of Fe-Ni-Cu sulphides Co, Au and PGE associated with cumulus ultramafic zone of ophiolite
complex is discovered in Sebuku Island through exploration program of PT. Sebuku Iron Lateritic Ores (SILO). Occurrence
of mineralization is dictated by normal ultramafic rock but within complex structural due to tectonic events in the island to
produce specific mineralization styles. At this stage, the styles are considered to be based on phase of deposition. Various
phases of deposition are analyzed through: (i) major and trace elements geochemical data derived from XRF, ICP-MS and
(ii) petrography-mineragraphy by transmited light microscopy, SEM and QEMSCAN. Phases of deposition are interpreted
as follow: (i) Cr-Fe oxide mineralization in serpentinite, (ii)Fe-Ni-Cu sulphide mineralization in cumulus ultramafic (iii)
enrichments through metasomatism and hydrothermal, and (iv) enrichments through latererisation due to weathering.
Trace metals are thought to be associated with the second mineralization phase.
INTRODUCTION
Concession of PT Sebuku Iron Lateritic Ores (SILO),
approximately 11,000 ha area, occurs in the North East
portion of Sebuku Island. SILO has produced iron lateritic
ores since 2004 and in construction period to produce
direct reduction iron and reduced nickel from detrital iron
and Fe-Ni bearing laterite ores. Since 4th quarter of 2014,
SILOs exploration and mining development aims to
explore mineral potential in laterite and primary
ultramafic complex unit.
Sebuku island is a small island located in SE corner of
South East Kalimantan, approximately 5 kms toward east
of Laut island. The island is administered by Kotabaru
Regional District within South Kalimantan. The easiest
access to the island from Jakarta, the capital city of
Indonesia is through one and a half hours flight to
Banjarmasin, followed by half an hour flight to Kotabaru
and followed by one hour of speedboat from kotabaru port
to Tanjung Nusantara port or Sungai Bali port.
GEOLOGY OF SEBUKU ISLAND
Geology of the area (Fig. 1) is covered by the youngest
quartenary unconsolidated sediment materials in most of
north west part and further south east by a coal bearing
sediments of Eocene-age Tanjung Formation and the rest
mostly occupies by ophiolite complex unit includes
pelagic sediment, gabbro dykes, cumulus ultramafic and
detrital iron ore unit, ages late Triassic to early
Cretaceous, except, to some extent, North South trending
of volcanoclastic of late Cretaceous Haruyan formation in
the central part of the island. Throughout pre Tertier,
several plutonic rocks intruded the units include gabbro,
diorite and andesite plutons.
Fig. 3 Geology Map of SILO Concession
Due to thick laterisation, limited accessible bedrock can

be identified during surface mapping. Due to the same
reason, soil sampling also incapable to provide useful
information. Also, limited active stream in the concession
make stream sediment geochemical sampling cannot
delineate prospect area. Best exposed lithology is within
pit perimeter although it also depends on Fe grade where
depth of excavation is limited to top saprolitic material
only. Second best bedrock observation is from drill core
which design to target 3m below enriched saprolite.
TECTONIC SETTING AND STRUCTURES
FROM AIRBORNE GEOPHYSIC
Most of authors of early tectonic studies in the regions
includes Koesumadinata, R. P. et. al., 1989; Van der
Weerd et. al. 1992; Setyana et. al., 1999; Wilson, M.E.J.
et. al.,1999; Moss, S.J., et. al., 2000; Hall, R., 2009 and so
106
PROCEEDINGS
on, agreed that Sebuku island is part of Laut Island ridge,

an oceanic crust material at convergent boundary to
Sundaland in NW and compressed toward WNW by
Patenoster Platform, a microcontinent, from SE. However,
disagreement is still exist, notably, whether the ridge is an
exposed sub lithosperic continental material or there ridge
is an obducted material as part of accretionary wedge of
subduction zone. Nonetheless, this tectonic activity forms
geology and structure of Sebuku Island.
Airborne Magnetic data carried out by the last quarter
of 2015. Interpretation from this data concludes an
identification of NNE major structural trend below 400m.
Structural interpretation above 400m influences by brittle
origin of serpentinised ultramafic, therefore incapable to
delineate useful trend. The magnetic data also provides
approximate location of intrusive bodies (dykes) and trend
of potentially shallow high magnetic anomaly. This
information was used to locate first potential target for
primary mineralization (Figure 2).
also altered to develop serpentine assemblage and or

soapstone/talc assemblage, due to gabbro dyke intrusion
cross cut this ultramafic to develop such low grade
metamorphism.
Fig. 5 Ultramafic and sulphide - oxide minerals
Petrology and Mineralogy of hostrock and the

intrusion, magmatic sulphide and other potentials were
examined by Prof. Dr. D. Pirrie in Camborne School of
Mines, using optical transmitted light microscopy, then
carbon coated for scanning electron microscopy in attempt
to obtain better quantitative information and furthermore,
studied by QEMscan with fieldscan method to obtain
higher degree of quantitative as well as to preserve
original fabric and texture of petrology and mineralogy.
Result of the study reveals that protolith (Figure 3) is
ultramafic and mostly altered to serpentinite with minor to
talc carbonate assemblage and some further altered to
tremolite. Younger phases of serpentine and chlorite also
are seen crosscutting older mineral in vein. Fe stringer is
developed along serpentine grain boundaries; euhedral
opaque spinel is disseminated throughout this rock and
sometimes zone out to magnetite or totally replaced by Fe
oxide. Minor Ni sulphides are disseminated throughout
sample and tend to be higher concentrate in association
with Fe oxide rather than in serpentinite.
Mineralogy of the massive sulphide is represented in
Figure 4. The massive sulphides consist of intergrown
pyrite, pyrrhotite, chalcopyrite, pentlandite, Copentlandite, cobaltite with inclusion of gold. Pd is the only
PGE exist in the sample even in low concentration. The
sulphides fracture fill and breccias matrices fill in
serpentine and chlorite of ultramafic origin which has
remnant euhedral chromite and fe-oxide magnetite
hematite.
Fig.4 Target Area Generated from Geophysic
FE-CU-NI SULPHIDE MINERALIZATION

First drillhole intercepts massive sulphide (MS)
mineralization is SDD0009 at the depth of 50.80 - 52.65
m. Others intercept various concentration of disseminated
(DS) and also massive sulphide mineralization ( up to
5m). Ni sulphide occurs as pentlandite (pn). In massive
sulphides, pentlandite are always in association with
magnetite (mag), pyrrhotite (pyr) with smaller amounts
pyrite (py), chalcopyrite (cpy) and other trace unidentified
minerals. All massive sulphides occur in fine grain and
euhedral form, as belbs-veins-stockworks to or
cut/scavange fractures/sheets of serpentinised ultramafic
host rock. Dissiminated sulphide exists as lower
concentration sulphide minerals ranging 1-3% up to 810% sulphide as crystal grain (replacing cpx? or Cr?) in
serpentinised ultramafic.
Host rock ultramafic majorly comprises dunite and
harzburgite with minor gabbro, anorthosite and troctolite
which occasionally develops thin stratiform layers at the
base, as indicator of in or near petrological moho
boundary between cumulate and tectonized ultramafic,
therefore, this unit is cumulus ultramafic. This cumulus is
107
PROCEEDINGS
Geochemistry of massive sulphide yields reasonable

fit to mineralogy discussed above. Outcome of the
elements can be seen in figure 6. Major elements i.e Fe,
MgO, SiO2, Al2O3, CaO, Na2O and K2O are still
reasonable for mafic-ultramafic classification without
acidic (contintental crust) influence and or hydrothermal
alteration. While sulphur is at reasonably high ~18% and
correspond to yield of Fe, Ni and Cu at 40%, 1.6% and
1,1% respectively. Co has positive correlation to Ni in
sulphide condition. High Ni sulphide is accompanied by
high Co thus Ni/Co ratio decrease. This relationship is
expected in Co-pentlandite. In SEM, other Co mineral,
Co-arsenides(cobaltite) is the only mineral is capability to
attach Au in its lattice. No anomalous PGE in this sample.
Fig 6 Massive Sulphides in thin section, SEM and QEMscan
Schulz, K. J. et. al., 2010 recognised these mineralogies

is set of minerals comprises of chalcophile elements
which bounding chemically to construct sulphides,
arsenides minerals within magmatic sulphide deposit
environment.
Gabbro intrusion and other potentials within SILO i.e.
stratiform chromite, hydrothermal As-Sb, detrital iron and
lateritic Fe-Ni deposits are not covered in this paper due to
relevance to the topic.
Figure 8 Geochemistry of massive sulphide
DISCUSSION
Schulz, K. J. et. al., 2010, mentioned that at present
there is no economic magmatic sulphide deposit generated
in pure MORB and Ophiolite complex environment.
Further study is carried out especially to follow up
remnant metamorphism in talc carbonate assemblages in
attempt to discover potential partial melting plume
magmatism which responsible to magmatic sulphide
occurrence in SILO.
Figure 7 Scatter Plot of Host Rock
GEOCHEMICAL DATA
CONCLUSION
Geochemical data is generated by XRF method for

major elements and ICP-MS for trace elements includes
rare earth element (REE). Sun and McDonough, 1998
discrimination, as suggested by Dr. Ade Kadarusman, is
used to analyze petrochemisty of host rock and scatter
diagram is utilized to evaluate relationship of sulphide
cation. Petrochemistry application to host rock is limited
to gabbro due to alteration of ultramafic to serpentine and
talc carbonate assemblages.
REE plots indicate the unit is in cumulus maficultramafic. Plot of normalized to primitive mantle shows
all elements are enriched relatively and normalized to
normal mid oceanic ridge basalt shows majorly slightly
below reference value. The deficit to reference value of
MORB is considered acceptable. This deficit is thought to
be caused by volatiles (H2O, CO2, S etc) introduce to this
unit. Trace element scatter diagram is confirming data
shown in REE plots.
Sebuku Island confirms to have magmatic sulphide

potential. This potential is majorly developed by specific
tectonic setting and structure so that the massive sulphide
exist in specific structural trap even in normal MORBorigin ophiolite environment and in no obvious sulphur
source or sulphur saturated environment.
Advance petrologicalmineralogical tools and
geochemical tools are proved to assist SILO to conduct
problem solving map to optimize exploration program.
REFERENCES
Kivior, I, 2015, Internal Report: Airborne geophysic based on
ACM techniques
Pirrie, D., 2016, Internal Report : Mineralogical Report
Schulz,K.J et. al.,2010, Magmatic Sulfide-Rich Nickel-Copper
Deposits Related to Picrite and (or) Tholeiitic Basalt DikeSill Complexes: A Preliminary Deposit Model
108
PROCEEDINGS
Geologists of SILO, 2014-2016; Various reports, weekly,

monthly, project and special
assistance in petrochemistry and Prof Dr Duncan Pirrie for

advance study in petrography and mineralogy.
ACKNOWLEDGEMENTS
Special thanks to SILO geologist who done most of
hard works in the field, Dr. Ade Kadarusman for
109
PROCEEDINGS
Investigation on Scandium and REE potential in Sulawesi, Indonesia: a

geochemical constraint
Adi Maulana1, Kenzo Sanematsu2, Masayuki Sakakibara3
1
Geology Department, Hasanuddin University, Makassar, Indonesia

2
Advanced Institute of Science and Technology, Tsukuba, Japan
3
Graduate School of Science and Engineering, Ehime University
Corresponding author: adi-maulana@unhas.ac.id
ABSTRACT
Scandium (Sc) and rare earth elements (REE) are an important element in the green technology application. An
investigation on the scandium and rare earth element potential in Sulawesi was conducted using geochemical data. Sc is
concentrated in limonite layer of the Soroako ultramafic rocks as a result of Fe3+ site substitution of mafic minerals
(pyroxene, amphibole, etc) during laterization processes. REE are enriched in association with clay mineral in B horizon
from heavily weathered granitic rocks in Palu and Masamba, suggesting the possibility of ion-adsorption style
mineralization. The lateritic soil of the ultramafic rocks and the weathered crust of the granitic rocks in Sulawesi could be
the potential source for scandium and rare earth elements, respectively.
Keywords : Scandium, rare earth element, Sulawesi, Indonesia.
show decreasing content of SiO2 and MgO, and increasing
trend of Fe2O3 and Sc from bedrock to limonite layer.
SiO2/MgO ratio and Sc show a slightly increasing trend
from bedrock to soft saprolite but significantly enriched in
yellow limonite layer. Overall, West Block weathering
profiles shows similar pattern to Petea profile. SiO2 and
MgO are significantly decreased whereas Fe2O3, Sc and
SiO2/MgO ratio are enriched toward the upper part of the
weathering profile. These two variations suggest the
enrichmentment of Sc in the limonite horizon as shown in
profile pattern.
Sc occurrence in Sulawesi was firstly reported by
Kadarusman et al. (2004) and Maulana et al. (2015) from
bulk rock composition of ultramafic rocks. It is reported
that Sc content is abundant in pyroxenite from Bantimala
Complex, South Sulawesi (Maulana et al., 2015). Wholerock geochemical data of the laterites from Soroako
ultramafic rocks suggest that Sc is likely to exist in Fe
oxides and pyroxene-rich bedrock (harzburgite in
composition). Notably, Sc tends to be enriched in limonite
layer. Sc is unlikely to be adsorbed on minerals and
amorphous materials in the laterites. Sc is more distributed
in pyroxene rich bedrock than in olivine since coefficient
value of scandium is hosted in orthopyroxene and
clinopyroxene as listed in Table 1. The enrichment of Sc
will be mainly concentrated in the weathering product of
pyroxene-rich bedrock, particularly in limonite layer with
low to medium Ni-content. Based on this, large potential
of Sc resources is expected in Sorowako because of large
weathered pyroxene-rich ultramafic rocks distribution as a
source of nickel (Ni)-laterite deposit.
INTRODUCTION
Scandium (Sc) and rare earth elements (REEs) are an
important element in the green technology application.
An investigation on the scandium and rare earth element
potential in Sulawesi was conducted using geochemical
data. Sc is concentrated in limonite layer of the Soroako
ultramafic rocks as a result of Fe3+ site substitution of
mafic minerals (pyroxene, amphibole, etc) during
laterization processes. REE are enriched in a s s o c ia t io n
w it h c la y m i ne r a l in B ho r izo n fr o m heavily
weathered granitic rocks in Palu and Masamba,
suggesting the possibility of ion-adsorption style
mineralization. The lateritic soil of the ultramafic rocks
and the weat hered crust of t he granitic rocks i n
S u l a w e s i could be the potential source of scandium
and rare earth elements, respectively.
DATA AND METHOD
The weathered ultramafic rocks were taken from
Soroako whilst granitic rocks samples were taken from
Palu and Masamba area. They were analyzed for
concentration of major elements using X-ray fluorescence
spectrometer (XRF) RIGAKU RINT-300 both in Advance
Institute of Science and Technology (AIST) Laboratory,
Tsukuba and in the Department of Earth Resources
Engineering, Kyushu University. Rare earth element and
trace element including scandium composition for both
samples groups were determined using ICP-MS following
lithium metaborate/ tetraborate fusion and nitric acid total
digestion at ALS Mineral, North Vancouver, Canada.
Geochemistry of REE in weathered granitic rocks
Geochemistry of Sc in weathering profile of ultramafic

rocks
Generally, the weathered crusts show higher

concentration of total REE and REE + Y compare to their
parent rocks as reported by Maulana et al (2014). This
results indicate that the weathered crust were enriched in
REE, consistent with the enrichment of REE in weathered
granitic crust which are reported from many areas
(Ishihara et al, 2008; Imai et al., 2013). However, Palu
weathered crust only shows a small enrichment of REE
compares to Mamasa weathered crust. The total REE
contents of the Mamasa weathered crusts range from 58 to
Most of ultramafic sequences in Soroako area have

been intensively weathered producing a lateritic profile
(Achmad, 2006). The profile consists mainly of bedrock,
saprolite, limonite and top soil layers showing various
thicknesses. Bedrocks are classified as harzburgite in
composition. The bedrock in Petea profile contains of 40.8
wt% SiO2, 34.8 wt% of MgO, 8.7 wt% Fe2O3, 0.2 wt% Ni,
9.8 ppm Sc. In relation to this, the weathering profile
110
PROCEEDINGS
Fig. 1. Granitic rocks, ultramafic rocks and lateritic Ni deposit

distribution in Sulawesi Island. Inset figure show ultramafic and
lateritic Ni deposit distribution in Philippines which have been
exploited for Sc (Granitic rocks distribution is from Maulana
et.al, 2013; ultramafic rocks distribution from Kadarusman et al.
2004; Ni deposit in Philippines is from Yumul, 2007).
552 ppm (322 ppm on average) whereas those of fresh

(parent) rocks range from 20 to 356 ppm with average of
198 ppm (sample MA-38 and MA-43B which have low
SiO2 content were not used in average calculation). The
LREE content of the weathered samples range from 46
to 488 ppm whereas the total of HREE content ranges
from 10 to 64 ppm. La and Ce are the most abundant
elements in all the samples. In contrast, the REE in Palu
weathered crusts only show a very small enrichment
(approximately 20%) compare to their parent rocks. The
REE in Palu weathered crust ranges from 124 to 314
ppm (223 ppm in average) whereas the REE in the
parent rock range from 196 to 251 (200 ppm in average).
The total REE+ Y in the weathered crust range from 220
to 337 ppm and 198 to 267 ppm in the parent rocks. The
enrichment of REE occurred mainly in B-2 horizon, where
the total REE is up to 314 ppm (REE + Y = 337 ppm).
The granitic rocks in Sulawesi are heavily weathered;
for example the Polewali and Mamasa areas (Maulana et
al., 2014). The weathering profile shows a thickness of 2
to 5 meters consist of at least of 2 horizons, A and B (Fig.
2b). REE content in bedrock of granitic rocks in Sulawesi
range from 110 250 ppm and enriched in weathered
profile by 1.5 to 2 times (Maulana et al., 2014). This
enrichment suggests the possibility of ion-adsorption type
deposit of REE in weathered granitic rocks from Sulawesi.
Maulana et al. (2014) reported that close association
between clay minerals such as montmorillonite and
kaolinite and REE enrichment are noted in Sulawesi. The
REE occurrence in Sulawesi is further supported by the Itype granitic rocks domination (Maulana et al., 2016).
The existences of Sc and REE at these concentration
levels in weathered ultramafic and granitic rocks have
significant implications in studying these elements. This
raised the economic interest as Sc and REE elements tend
to concentrate in weathering profile of ultramafic and
granitic rocks, and also because Sc is notably fixed in Fe
rich mineral and REE which were hosted in weatheringproduct mineral (clay mineral). Thus, it is shown that both
Fe rich mineral (pyroxene) may be significantly host for
Sc in weathered ultramafic rock and clay mineral may
adsorb REE in weathering processes of granitic
rocks.From the economic point of view, the high levels
and uniformity distribution of Sc as well as REE
suggested by our study indicate that the limonite layer in
weathering profile of ultramafic rocks and B horizon in
weathered crust of granitic rocks may be a valuable source
for Sc and REE, respectively.
CONCLUSIONS
Sc-bearing laterite Ni deposit in Sulawesi could be a
predominant Sc resources in the near future. Sc is likely
to substitute Fe3+ site of mafic minerals in weathering
product of pyroxene-rich ultramafic bedrock; it this case,
further study is required.
REE resources in Sulawesi can be extracted from ionadsorption type deposit from heavily weathered I-type
granitic rocks in Sulawesi.
As weathered ultramafic and granitic rocks in Sulawesi
are widely distributed, it is expected that Sc and REE can
be economically extracted in the future.
Further detail study on the occurrence of these critical
metals (Sc and REE) therefore should be conducted
intensively in order to maximize the potential of these
materials for better development.
REFERENCES
Imai, A., Yonezu, K., Sanematsu, K., Ikuno, T., Ishida, S.,
Watanabe, K., Pisutha-Arnond, V., Nakapadungrat, S. and
Boosayasak, J. (2013). Rare earth element in
hydrothermally altered granitic rocks in the Ranong and
Takua Tin-Field, Southern Thailand. Resource Geology
63(1) 84-98.
Ishihara, S., Hua, R., Hoshino, M., and Murakami, H. (2008).
REE abundance and REE minerals in Granitic Rocks in the
Nanling Range, Jiangxi Province, Southern China, and
Generation of the REE-rich Weathered Crusts Deposits,
Resource Geology, 58 (4), 373 - 401
Kadarusman, A., Miyashita, S., Maruyama, S., Parkinson, C.D.
and Ishikawa, A. (2004). Petrology, geochemistry and
paleogeographic reconstruction of the East Sulawesi
Ophiolite, Indonesia. Tectonophysics, 392, 55-83.
Maulana, A. Christy, A.G. & Ellis, D.J. (2015). Petrology,
geochemistry and tectonic significance of serpentinized
ultramafic rocks from the South Arm of Sulawesi. Chemie
der Erde, 75(1). 73-88
Maulana, A. Imai, A., Van Leeuwen, T., Watanabe, K., Yonezu,
K, Nakano, T., Boyce, A., Page, L. and Schersten, A.
(2016). Origin and geodynamic setting of Late Cenozoic
granitoid in Sulawesi, Indonesia. Journal of Asian Earth
Sciences. 124, 102-125.
Maulana, A., Yonezu, K. And Watanabe, K. (2014).
Geochemistry of rare earth elements (REE) in the weathered
crusts from the granitic rocks in Sulawesi Island, Indonesia.
Journal of Earth Science. 25(3), 460-472.
Yumul, G.P.Jr. (2007). Westward younging disposition of
Philippine ophiolite and its implication for arc evolution.
Island Arc 16, 306-317.
ACKNOWLEDGEMENTS
We would like to express our sincere gratitude to
KLN DIKTI research scheme 2015 for field work budget
and JICA C-BEST UNHAS Research Grant 2015. PT
Vale Indonesia, Soroako is greatly acknowledged for
their facility and permit during field work.
111
PROCEEDINGS
Uranium Exploration in Sulawesi

I G. Sukadana1, and H. Syaeful1
Centre for Nuclear Mineral Technology, National Nuclear Energy Agency, INDONESIA.
e-mail: sukadana@batan.go.id
ABSTRACT
Uranium exploration was conducted since 1978 covering almost all of Sulawesi Area including Banggai and Sula Islands.
Granitic rocks and aluvial placer deposit was targeted as the source and host rock of radioactive minerals deposits in
Sulawesi. Alkaline rock also became exploration target of radioactive minerals exploration recently. The methode that was
applied in uranium exploration are combine the geological data, radioactivity measurement and geochemical analysis.
Some area has high content of uraium, such as Masamba, Pasangkayu, Banggai, Sula, Barru and Mamuju. Uranium
content in exploration area usualy not to high except in Mamuju. Uranium minerals can be occurs in rocks as its own
minerals, as a substitute element in accessory minerals, adsorbed on Fe and Fe-Ti hydroxides-oxides and clay minerals
adsorbed by organic matter, in crystal defects of major rock forming minerals, and dissolved in geologic fluids and fluid
inclusions. Uranium also generally low concentrations (1 to several ppm), most of the uranium is substituted in the
structure of one or several accessory minerals such as apatite, zircon, monazite, xenotime, and titanite, which represent
refractory uranium sources for mobilization by most types of geologic fluids. Radioacive minerals was found in Sulawesi
such as monazite, allanite, thorianite and titanite group (davidite). Exploration of uranium in Mamuju area is associaeted
with rare earth elements (REE). Development of uranium minerals exploration was focused in Mamuju Area West Sulawesi.
Key words: Uranium, exploration, Granitic, Alkaline rocks, Sulawesi.
INTRODUCTION
Uranium exploration in Sulawesi has been conducted
from 1978 to 1984 (Ngadenin et al. 2010). The activity
was slow down in 1984 and begin in 2012 in Mamuju
Area. Granitic rocks, acid volcanic rocks and aluvial
deposits has been the target of uranium exploration.
Uranium minerals naturaly can be formed in plutonic,
metasomatic,
hydrothermal,
basinal
diagenetic,
metamorphic, and volcanic to sedimentary and superficial
environments with the calcretes (Bruneton and Cuney
2016). The aim of the research are to define the
distribution of the potential rocks and potential deposits in
Sulawesi base on uranium exploration data.
Sulawesi can be divided into four (4) tectonic
provinces namely (1) the Western and North Sulawesi
Pluto-Volcanic Arc, (2) the Central Sulawesi
Metamorphic Belt, (3) the East Sulawesi Ophiolite Belt
and (4) the Banggai-Sula and Tukang Besi continental
fragments (Maulana et al. 2013). Uranium exploration was
conducted in all area except the East Sulawesi Ophiolite.
The exploration step are various from early/previous
study, general prospection and detailed prospection
(Figure 1).
Preliminary study conducting by literature and
previous report and general geology and tectonic setting
of some area. The results of preliminary study will follow
up by general prospection completed by geological
information, regional data colection (geochemistry and
rock sampling) also completed by radiometric mapping
with regional scale. That methode will be resulting the
radiometric map and uranium content from every sampel.
The result of this step will be follow up by detailed
prospection.
Fig. 1: Exploration step and location in Sulawesi.
Geological setting is one of important aspect in the

uranium exploration. Uranium distribution in the Earths
crust that commonly persisted through long periods of
time, and through a combination of orogenic,
metamorphic, and sedimentary processes produced rocks
with enriched uranium contents. The initial enriched
uranium domain was successively remobilized and
concentrated into new enrichments of one or more
magnitudes above normal background forming uranium
112
PROCEEDINGS
ore deposits(Bruneton and Cuney 2016). These process

will be resulting the potential uranium deposits area.
Rock Type
Average
o
(c/s)
SPP 2NF
METHOD AND DATA
Late
Tertiary
Sediment, and part of
Quartenary sea sediment
Midle
Late
2 Tertiary Volcanic
Preliminary study, general and detailed prospection

was conducted in several location in Sulawesi from the
North to the South. Radiometric field measurement
conducted by gamma suveyor SPP2NF to measure the
intensity of gamma radiation and RS 125 to measure
radiation dose rate, potasium (K), Thorium (Th) and
Uranium (U) content. Laboratory analisis conducted in
using UV-Vis Spectrometri.
Northwest area of Sulawesi covered Pasangkayu was
conducted on a triple contact between metamorphic
(Latimojong formation/Kls), granitic (Batuan Terobosan
unit/Tmpi)
and
sedimentary
rock
(Lariang
Formation/Tmpl). There are no potensial area of uranium
anomaly in this area event there are posibility from
lithology and stratigraphy aspect. Other location in
Northwest of Sulawesi is Bangkir Area, Teluk Tambu,
Teluk Dondo, Tanjung Tibo and Tindaki village has low
content of anomaly with uranium content > 1 ppmU.
In Central Sulawesi, uranium exploration focused in
Masamba Area. The exploration located on the north of
Boni bay (Teluk Boni). The objective of this area are
Tertiary Kabuno Granite (Tpkg) consist with granite,
schists with pegmatitic. The result of this area has average
of uranium content 5-20 ppmU. Some anomalies reach
100 ppmU and very consistent, with large area arround
300 m x 200 m. Kulawi Area is other exploration location
in Central Sulawesi. The target of these area is granite and
metamorphic formation along Kulawi river. Lithology of
this area consist migmatite (Mg), Matamorphic (M),
Intrusive granite (Td), volcanic material (Hga) and
sedimentary Formation (Sd). The value of uranium
content in the stream sediment have significan anomaly
about 3 6 ppmU.
Banggai archipelago It comprises of four islands:
Peleng, Banggai, Labobo and Bangkurung resting at a
close distance from the eastern coast of Sulawesi. The
target in this area mostly the Banggai granite and, as
secondary objective, the sedimentary formations of the
Bobang formation. The only anomalies detected, both
geochemical and radiometric, coincide almost perfectly
with the extension of the Banggai granite on these islands.
Geochemical survey shows the very good correlation
between the stream sediment results that are coinciding
with the extension of the granite.
Sula Island exploration target has objective a variety of
terrains, metamorphic, granitic and sedimentary.
Radiometric results are very high readings on the Banggai
granite with a background close to 200 c/s SPP2.
Geochemical results: also carried on the same surface.
Weak values on the sedimentary formations ( 0,5 ppmU
0,7 ppmU). There are much higher values on the Bonggai
granite (2 ppmU 10 ppmU). The contact between granite
and sedimentary rocks, outlined by the presence of a
conglomerate has not been seen and is not located with
radiometric or geochemistry measurements. Several result
uranium exploration in Sulawesi are explain in Tabel 1
and the location in Figure 2.
Felsic
Intermediete Tertiary
Intrution
4
Early Tertiary
Cretaceous
Clastic
Sediment
5
Triassic
Metamorphic
Anomal
y
(c/s)
SPP 2NF
40
100
100
300
125
300
40
100
75
150
Anomaly
Location
Pasangkayu
,
Polewali
Mamasa
Masamba
Hulu, Polewali
Mamasa
Bantimala and
Maros Gowa
Bangkir,
Sabang
and
Masamba
Masamba
Hulu
Masamba
and Kulawi
Fig. 2: Geological map (Sukamto R., 1975) completed by

anomaly of radioactivity and uranium content .
Recenly, starting from 2012, uranium exploration

focused in Mamuju Area West Sulawesi. Mamuju has
highest radiation dose rate in Indonesia (Iskandar, et al.,
2014) The radiation value due to its Naturally Occurring
Radioactive Material (NORM), has been identified in the
area of Adang (Syaeful, et al., 2014). These volcanic rocks
are composed of ponolite to dacite rock, with ultrapotassic affinity, formed in active continental margin
(ACM)(Sukadana et al. 2015). Distribution of high
radiation in this area are controled by volcanic rocks
distributions(Indrastomo et al. 2016).
The result of uranium and thorium analysis several
sample from Mamuju has high content of uranium and
thorium. The mineralization type of uranium
mineralization is volcanic related of uranium deposit type.
The result of UV-Vis analysis from several sampel in
Mamuju as shown in Tabel 2.
Table 1. The sumary of Uranium exploration Result
113
PROCEEDINGS
Table 2. Geochemistry analysis result of several Mamuju

sample.
N
Sample Code
U (ppm)
Th ( ppm)
1
2
3
4
5
6
7
8
9
1
MJU/03
MJU/06
MJU/07
MJU/09
MJU/20
MJU/21
MJU/30
MJU/45
MJU/56
MJU/61
1104,0
1729,0
2291,1
1662,0
426,7
366,1
224,4
1354,0
713,9
1113,0
216,8
133,7
209,7
518,4
13.240,0
9919,0
163,7
814,4
618,3
212,5
series. Its explained by assimilation process between

magma and crustal material that containing a various
quantity of reduced C-and S-bearing sediments (Maulana
et al. 2013). These two magmatic series are considered to
represent different redox condition, source rock
composition, and crystallization and emplacement history
These are mostly I-type, with the CAK series also
including a late phase S-type (Maulana et al. 2016). The
granitic rock that has high content of uranium are S-Type
granite.
The result of recently uranium exploration in
Mamuju, West Sulawesi was changed the mindset of
uranium exploration in Indonesia. Usualy the target of
uranium exploration was felsic plutonic and volcanic
rocks. The formation of Adang volcanic composed by
feldspathoid lava rock, pyroclastic, tuffities with
intermediete to basaltic composition (Sukadana, et al.
2015). These volcanic products are representative of
orogenic magmas emplaced in a subduction context, with
metasomatized mantle sources conforming to those of
converging continental margin basalts (Conte et al. 2016).
Edifice of volcanics rock, the morphology, characteristics
and petrochemical composition of volcanic outcrops
recognized in the submarine portions allowed to enlarge
the knowledge on the development of the volcanic activity
in the area(Romagnoli et al. 2013). The increase of
uranium elements in volcanics rock are corelated to the
increase of K2O(Burwash and Cavell 1978).
The uranium exploration in Mamuju Area was shown
the impact of alkalinity of rocks with the content of
uranium. This model can adobe to develop of uranium
exploration methode in the future.
The samples are from several area in Mamuju, such

as Salunangka (Sample MJU 03), Botteng (Sample MJU
06, MJU 07, MJU 09), Hulu Mamuju (Sample MJU 20,
MJU 21), Takandeang (Sample MJU 56, MJU 61), and
Ahu (Sampel 30, MJU 45). The distribution of the
sampling area are shown in Figure 3.
CONCLUSION
Uranium exploration in Sulawesi was conducted since
1978 1984 with the object of uranium exploration area
are granitic, metamorphic and its related sedimentary
rocks. Recently uranium exploration was expans to
alkaline volcanic rocks such as in Mamuju Area. The
significant content of uranium was found in Adang
volcanic rock in several places in Mamuju, West
Sulawesi.
Figure 3: Iso Uranium Map as the result of RS 125 field

measurement
DISCUSSION
ACKNOWLEDGEMENTS
Early stage of exploration in Sulawesi, was focused in

granitic rocks and its related sedimentary and
metamorphic rocks. Uranium deposits resulting solely
from magmatic processes such as partial melting or
fractional crystallization are rare, in most cases, uranium
is initially mobilized from igneous rocks by hydrothermal
and/or surficial fluid(Ballouard et al. 2017). Anomalous of
uranium content and rock radioactivity found in Triassic
metamorphic rocks group, late Cretaceous to early
Tertiary clastic sedimentary rocks group, Tertiary acid to
intermediate intrusive rocks, middle to late Tertiary
volcanic rocks and late Tertiary sedimentary rocks. These
distribution of uranium are related to the series of granitic
Ilmenite series rock type in Sulawesi. The granitic rocks
in the northern part belong to magnetite-series whereas
those in the southern part generally belong to ilmenite-
We are apreciated to Centre for Nuclear Mineral

Technology, Tational Atomic Energy Agency, and college
who assit data collection in the field.
REFERENCES
Ballouard, C., M. Poujol, P. Boulvais, J. Mercadier, R. Tartse,
T. Venneman, E. Deloule, et al. 2017. Ore Geology Reviews
80.
Elsevier
Inc.:
30931.
doi:10.1016/j.oregeorev.2016.06.034.
Bruneton, Patrice, and Michel Cuney. 2016. Uranium for
Nuclear Power. Elsevier Ltd. doi:10.1016/B978-0-08100307-7.00002-8.
Burwash, R A, and P A Cavell. 1978. Contributions to
Mineralogy and Petrology 66: 24350.
Conte, A M, C Perinelli, G Bianchini, C Natali, E Martorelli, F L
Chiocci, Cnr-istituto Geoscienze, et al. 2016. Journal of
114
PROCEEDINGS
Volcanology and Geothermal Research. Elsevier B.V.

doi:10.1016/j.jvolgeores.2016.08.005.
Indrastomo, Frederikus Dian, I Gde Sukadana, Asep Saepuloh,
Agus Handoyo Harsolumakso, and Dhatu Kamajati. 2016.
Eksplorium
36
(2):
7188.
URL:
http://jurnal.batan.go.id/index.php/eksplorium/article/view/2
632.
Iskandar, Dadong, Syarbaini, and Kusdiana. 2014. Map of
Environmental Gamma Dose Rate of Indonesia. Jakarta:
National Nuclear Energy Agency.
Maulana, Adi, Akira Imai, Theo Van Leeuwen, Koichiro
Watanabe, Takanori Nakano, Adrian Boyce, Laurence Page,
and Anders Schersten. 2016. JOURNAL OF ASIAN EARTH
SCIENCES. Elsevier Ltd. doi:10.1016/j.jseaes.2016.04.018.
Maulana, Adi, Koichiro Watanabe, Akira Imai, and Kotaro

Yonezu. 2013. 6: 5057. doi:10.1016/j.proeps.2013.01.007.
Ngadenin, Heri Syaeful, P Widito, and Agus Sutriyono. 2010.
Urania 16 (No. 2 April 2010): 47104.
Romagnoli, C, D Casalbore, A Bosman, R Braga, and F L
Chiocci. 2013. Marine Geology 338. Elsevier B.V.: 3045.
doi:10.1016/j.margeo.2012.12.002.
Sukadana, I Gde, Agung Harijoko, and Lucas Donny Setijadji.
2015.
Eksplorium
36
(1):
3144.
http://jurnal.batan.go.id/index.php/eksplorium/article/view/2
312.
Syaeful, Heri, I Gde. Sukadana, and Agus Sumaryanto. 2014.
Atom
Indonesia
40
(1):
3339.
http://aij.batan.go.id/index.php/aij/article/view/263.
115
PROCEEDINGS
Eksplorasi Logam Tanah Jarang Di Jalur Timah Indnesia

Setiawan Raharjo
Kepala Unit Eksplorasi PT Timah (Persero) Tbk.
ABSTRACT
The rare earth element (REE) are contained in almost all types of igneous rock (igneous rocks ofacid, intermediate, alkaline
and ultramafic). In The Indonesia tin belt zone has dominated by igneous rocks of acid (Granitic), both S-type granites and
granite I-Type. The presence of rare earth metals in acid rock relatively smaller than ultramafic igneous rocks. REE in
Indonesia tin belt zone form of mineral monazite, xenotime and zircon. The existence of these minerals associated with the
deposits of tin, either in the form of primary deposits and alluvial deposits. These minerals generally have low grade but
relatively widely spread.In some places has high grade of REE but the levels are still below the BEG of tin mining.
Therefore, until now PT Timah conducting exploration and mining together with the explorationand mining of tin. The
spread of the primary tin and REE deposits is controlled by Granitic rock intrusion, so that their exploration stage are:
geological mapping (1), geochemical surveys (2), geophysical surveys (magnetic, gravity and resistivity-IP) (3) and drilling
(4). Spreading sediment alluvial tin and REE controlled by the source rock, the process of weathering and erosion,
sedimentation processes and traps (valleys, rivers, and beaches). Therefore, in conducting exploration activities deposition
of tin and REE placer carried out in phases: Conducting a survey where the source rock (granite carrier tin) (1), mapping
the topography and the face of the seabed (bathymetry) (2), seismic surveys (3) and drilling (4). In addition, there are
certainly activities sample preparation, laboratory analysis, databaseintegration and resource estimate.
KEYWORDS: exploration, deposits of REE, resources estimation of minerals
radioaktif untuk lebih berkembang. Apabila tidak ada
perubahan dalam bidang industry yang mengakibatkan
penurunan konsumsiunsur tanah jarang, Alonso et al
(web.MIT.edu, 2012), memproyeksikan kenaikan
permintaanglobal per tahun sebesar 5,6% sebagai batas
paling atas. Tinjauan potensi ini dibuat untuk
mengestimasi berapa besar sumberdaya REE yang
terdapat diwilayah IUP PT Timah, sebagai dasar-dasar
penentuan kebijakan strategis terkait eksplorasi mengenai
REE guna menunjang pengolahan RE mineral dalam skala
industri di masa mendatang.
Fluktuasi harga REO (Rare Earth Oxide) yang saat ini
memang menunjukkan grafik yang stagnan, namun
sempat mencapai harga tertinggi per kg, sebagai contoh
cerium oxide dengan harga jual$350/kg, menunjukkan
bahwa bahan galian tambang ini memiliki potensi pasar
yang cukup baik. Penurunan harga, sebagaimana komoditi
tambang lainnya yang mengalami penurunan termasuk
timah, namun dengan harga komoditi logam jarang
oksida, dapatlah dipandang menjadi salah satu alternatif
pundi pemasukan PT. Timah, Tbk kedepan tanpa
mengesampingkan core business sebagai perusahaan
penghasil logam timah. Adanya Feasibility Study yang
menggambarkan
flowchart
pengembangan
REE
diharapkan menjadi blueprint dalammemperhitungkan
nilai keekonomian serta optimalisasinya dalam eksploitasi
bahan galian, yangdiharapkan dapat diintegrasikan dengan
penambangan timah, yang dapat membantu dalam
meningkatkan nilai tambah REE yang berada dalam IUP
PT. Timah, Tbk.
PENDAHULUAN
Indonesia menjadi produsen yang signifikan untuk
sejumlah komoditi meliputi: batubara,tembaga, emas,
timah, dan nikel (Winzenried dan Halim, 2015: 1). Khusus
untuk timah, Indonesia bahkan mampu menduduki
peringkat negara produsen dan eksportir timah terbesar
kedua di dunia PT Timah (Persero) Tbk., 2015: 1) dengan
total produksi mencapai 27.431 ton pada tahun 2015.
Namun demikian harga timah terus merosot semenjak
2013; bahkan mencapai 16.186 USD/ton di tahun 2015
(P.T. Timah (Persero) Tbk., 2015: 13). Rendahnya harga
jual logam timah dapat disiasati dengan cara
mengeksploitasi komoditi ikutan timah lainnya. Menurut
Setijadji dkk. (2003) dan Szamalek dkk. (2013) komoditi
ikutan timah yang telah terbukti hadir adalah unsur tanah
jarang (UTJ). UTJ hadir dalam wujud mineral-mineral
berat (xenotime dan monazite) yang bercampur dengan
plaser mineral berat (cassiterite) yangmengandung timah.
Sebagaimana diketahui bersama penggunaan REE
europium,gadolinium sangat luas dan erat kaitannya
dengan produk industri teknologi tinggi, seperti industri
komputer, telekomunikasi, nuklir, dan ruang angkasa. Di
masa mendatang diperkirakan penggunaannya akan
meluas, terutama REE tunggal, seperti neodymium,
samarium,, dan yttrium.
Kebutuhan akan logam tanah jarang (LTJ) seiring
dengan pertumbuhan industri menunjukkan trend yang
makin meningkat. Perkembangan industry chemical, metal
alloys, catalys menjadi angin segar bagi keberlangsungan
penggunaan REE. Cina dan Amerika Serikat telah sebagai
salah satu negara yang memiliki cadangan REE,
memanfaatkan celah ini sebagai importir tetap dengan
volume penjualan yang semakin meningkat. Kebijakan
reformasi penggunaan sumber energy dalam Konferensi
Perubahan Iklim Paris, mengakibatkan kebijakan green
energy menjadi program negera maju dan berkembang.
Sehingga optimalisasi pembangkit listrik bertenagakan
non fossil menjadi kesempatan unsur tanah jarang dan
116
PROCEEDINGS
Proyeksi Permintaan REE tahun 2010 s.d 2035
Tenggara yang oleh Cobbing dkk, 1992 disebut sebagai

Granite Tin Belt .
Sekarang diterima oleh kalangan geologiwan bahwa
inti dari SE Asia dirakit dari blok benua yang terpisah dari
Gondwana di Paleozoikum dan digabung dengan blok
Asia di Triassic. Beberapa fragmen dari blok-blok dirakit
rift basin dan dipisahkan dari Asia dan kemudian kembali
bergabung dengan bagian dari SE inti benua Asia tertua
sisi barat selama Mesozoikum. Ada semakin banyak bukti
bahwa fragmen Cathaysian / Asia membentuk kerak
benua Kalimantan Utara dan kerak lepas pantai di utara
Sarawak dan timur dari Vietnam. Sarawak, lepas pantai
Grounds Luconia-Berbahaya daerah, dan Palawan
termasuk bagian dari benua Asia tersebut. Ini mungkin
mewakili batas zona akresi di Asia-Pasifik, yang
merupakan tepian benua aktif sampai awal di Akhir Kapur
(Hall, 2014). B benua lainnya dibelah dari Australia dalam
Jurassic (SW Kalimantan, Sulawesi Tenggara Jawa-Barat,
Sulawesi Sabah-NW, South Sulawesi-Sumba), dan Woyla
intra-samudera busur Sumatera, dan ditambahkan ke
Sundaland pada Cretaceous. Setelah tabrakan blok
tersubduksi berhenti disekitar Sundaland pada awal Kapur
Akhir, dan dari sekitar 80 Ma sebagian besar Sundaland
adalah muncul, secara fisik terhubung ke Asia, tetapi
dipisahkan oleh lautan dalam dari India dan Australia.
Catatan harus diambil bahwa Pulau Bangka sepenuhnya
terletak di dalam Bentong - Raub Jahitan Zone (Hall,
2014). Secara implisit pun proses evolusi pergerakan jalur
subduksi diutarakan oleh Katili, 1972 yang secara tidak
langsung menunjukkan posisi kolisi di sepanjang jalur tin
belt tersebut.
Gambar 1. Prospek ekonomi REE tahun 2010 s.d 2035
Kemenerusan pasokan menjadi salah satu kunci utama.

Untuk memastikan kemenerusan pasokan, maka
keyakinan akan keberadaan sumberdaya/cadangan REE
harus menjadi prioritas,baik volume maupun kadar
ekonomis yang berada di dalam IUP. Timah, Tbk.
Setting Geologi Timah dan Mineral Ikutan Timah
Endapan timah aluvial sebagaimana tipikal komoditas
mineral lainya memiliki mineral-mineral lain sebagai
asosiasi mineral dan dibedakan atas dua tipe yakni mineral
berat (heavy minerals) dan mineral ringan (light
minerals). Mineral ringan bermakna sebagai mineral
dengan specific gravity < 2.5 gr/cc sedangkan mineral
berat adalah mineral dengan specific gravity > 2.5 gr/cc.
Jumlah mineral berat sangat dominan > 99.8% yang
sangat didominasi oleh mineral kuarsa sedangkan mineral
berat kurang lebih hanya sekitar 0.001 - 0.2 % saja yang
umumnya terdiri atas (Data PT Timah (Persero) Tbk);
Casiterite (SnO2), SG : 6.6 - 7
Hematite (Fe2O3), SG : 5.3
Ilmenit (FeTiO2) , SG : 4.5-5
Xenotime/Yttrium Phosphate (YPO4), SG :4.4 5.1
Monazite (Ce,La,Th,Nd,Y)PO4, SG : 4.6 5.7
Zircon (ZrSiO4), SG : 4.6 4.7
Marcasite (FeS2), SG 4.8
Rutile (TiO2), SG 4.2
Limonite (FeO3), SG 2.9 - 4.3
Siderite (FeCO3), SG 3.9
Biote (K(Fe,Mg)3 AlSiO3O10 (F,OH)2K, SG 2.9

Sehingga dengan kondisi seperti diatas maka endapan
mineral ikutan timah merupakan hasil proses mineralisasi
yang sama dengan endapan timah yakni dikontrol oleh
proses magmatisme granite pembawa timah.
Daerah operasional PT Timah (Persero) rbk berada di
Pulau Bangka, Belitung dan Perairan Karimun Kundur
yang merupakan bagian dari jalur metallogenic timah Asia
Gambar 2 Persebaran blok di Asia Tenggara. Terrain Sundaland

berumur Pre-Karbon merupakan modifikasi setelah Metcalfe
(1996, 2011a,b) dan Barber, dkk, (2005) (dalam Hall,2014).
Sumatra bagian barat, Burma Barat, dan Indochina- East Malaya
merupakan bagian dari Blok Cathaysian termasukEurasia pada
Paleozoikum. Sibumasu terakresi sepanjang Sutur Bentong-Raub
saat Trias (Hall, 2014).
Endapan timah itu sendiri terbentuk pada jalur granit

adalah sebagai akibat proses interaksi magmatisme
dengan kerak benua atau disebut dengan proses peleburan
kerak benua(crustal anatexis) sehingga terjadi pengkayaan
Sn selama terjadi fase tersebut.
117
PROCEEDINGS
Gambar 3. Proses terbentuknya granit berimah disepanjang jalur

tin belt
Karakteristik Granit Pembawa Timah

Sumber utama timah dunia adalah batuan beku
intrusif asam atau granitik atau granitoid Namun hanya
tipe batuan granitik tertentu saja yang dapat menghasilkan
bijih timah primer yang ekonomis sementara itu tiga
perempat produksi timah dunia berasal dari hanya enam
negara, tiga di antaranya berada di zona sabuk granit Asia
Tenggara. Dalam hal ini mengapa hanya ada beberapa
tempat saja provinsi metalogeni timah dunia menjadi
pertanyaan ahli geologi sekian lama (Lehmann, 1990;
Kesler dan Simon, 2015). Hal ini menunjukkan bahwa
timah hanya terbentuk pada tipe batuan granitik yang
sangat khusus Gasparon and Varne (1995) Hasil penelitian
Ishihara, 1981 men unjukkan bahwa tipikal granit
pembawa timah bercirikan hadirnya mineral ilmenit atau
disenut dengan ilmenite series granite atau ini paralel
dengan tipe granit S menurut Chappel and White, 1974.
Namun granit tipe I juga bisa menjadi granite pembawa
timah (ilmenite series) ketika magmatism peleburan
batuan beku mengalami proses reduksi. Sehingga dapat
diketahui bahwa karakter granit pembawa timah adalah
:Seri ilmenit (umumnya tipe S tapi beberapa tipe I)
1) Peralkalin
2) Sangat asam (benar-benar granit)
3) Mineral khas: ilmenit (bukan magnetit), biotit (bukan
hornblenda)
4) Sifat kemagnetan (magnetit susceptibility) rendah
5) Pada zona mineralisasi ditemukan mineral yang
sangat khas yaitu mika muskovit (greisen)
Gambar 5. Sebaran granit pembawa timah di Pulau Bangka
Proses Pembentukan Endapan Timah Placer dan

Mineral Asosiasinya
Syarat terbentuknya endapan timah aluvial adalah
sebagai berikut ;
6) Hadir sumber (Granit pembawa timah)
7) Proses pelapukan dan erosional - transportasi
8) Proses Sedimentasi
Untuk point pertama sudah dibahas secara panjang
lebar pada bagian sebelumnya, selanjutnya akan dibahas
untuk point kedua dan ketiga. Untuk syarat kedua sangat
dikontrol kuat oleh kondisi cuaca (climate) dan fluktuasi
muka air laur (sea level change). Tanpa hadirnya iklim
yang lembab (humid) maka proses pelapukan yang
intensif tidak akan terjadi serta proses pelapukanitu sendiri
tidak akan maksimal seandainya material yang akan
dilapukkan tidak berada pada suhu atmosfir (atau
tersingkap dipermukaan) yang dalam hal ini merupakan
peran turun naiknya muka air laut (sea level change,
Gambar 6).
Intensifikasi proses pelapukan terjadi pada jaman
Pleistosen dimana pada jaman tersebut terlihat bahwa
pada saat terjadi kondisi iklim dan sea level change yang
kondusif menjadi saat yang paling intensif terbentuknya
endapan placer (Gambar 8).
Gambar 6. Perubahan muka air laut dan iklim global sebagai

kontrol terbentuknya endapan timah aluvial
Gambar 4. Karakteristik granit pembawa timah
118
PROCEEDINGS
terlebih dahulu dari senyawa kompleks tersebut. Selama

ini telah diketahui lebih dari 100 jenis mineral RE, dan 14
jenis di antaranya diketahui mempunyai kandungan oksida
RE yang tinggi. Mineral RE tersebut dikelompokkan
dalam mineral karbonat, fospat, oksida, silikat, dan
fluorida. Mineral RE bastnaesit, monazite, xenotime dan
zircon paling banyak dijumpai di alam.
Mineral RE yang terdapat di jalur tin belt Asia
Tenggara khususnya pada wilayah IUP PT Timah
(Persero) Tbk. yaitu kelompok mineral fosfat terdiri dari
monazite dan xenotime. Selain itu terdapat juga zircon
yang tersebar luas di semua wilayah. Mineral-mineral
tersebut merupakan hasil sampingan dari penambangan
dan pengolahan timah aluvial (MIT :Mineral Ikutan
Timah) yang saat ini sebagian besar terakumulasi pada
tailing penambangan dan belum dimanfaatkan.
Gambar 7. Korelasi pengaruh sea level change-global climate

dengan terbentuk endapan timah
Karakteristik REE
REE dalam sistem periodik unsur berada dalam
golongan lanthanida yang terdiri atas 15 unsur (Gambar
8), mulai dari lanthanum (No atom 57) hingga lutetium
(No atom 71). Yttrium, scandium, dan thorium
dimasukkan
kedalam
kelompok
REE
dengan
pertimbangan kesamaan sifat. Unsur tanahjarang
mempunyai sifat reaktif tinggi terhadap airdan oksigen,
bentuk senyawa stabil dalam kondisioksida, titik leleh
relatif tinggi, serta sebagai bahanpenghantar panas yang
tinggi. REE bersifat tidak tergantikan. Hal ini disebabkan
sifatnya yang sangat khas, sehingga sampai saat ini, tidak
ada material lain yang mampu menggantikannya. Jika ada,
kemampuan yang dihasilkan tidak sebaik material REE.
Sifat REE yang digunakan sebagai material berteknologi
tinggi dan belum ada penggantinya, membuat logam tanah
jarang manjadi material yang vital dan mempunyai potensi
strategis. Berdasarkan variasi radius ion dan susunan
elektron, REE diklasifikasikan ke dalam dua subkelompok, yaitu :
1) REE-ringan, atau subkelompok cerium yang meliputi
lanthanum hingga europium
2) REE-berat, atau subkelompok yttrium yang meliputi
gadolinium hingga lutetium dan yttrium.
Gambar 9 Kandungan REE di dalam monazite Pulau Bangka

(presentasi BATAN, 2014)
Monazite((Ce,La,Y,Th)PO3)
merupakan
senyawa
fosfat-REE yang mengandung 50-70% oksida REE.
Monazite memiliki kandungan thorium yang cukup tinggi
sehingga bersifat radioaktif. Thorium memancarkan
radiasi tingkat rendah, dengan menggunakanselembar
kertas saja akan terhindar dari radiasi yang dipancarkan.
Monazite terbagi menjadi 3 tipe seperti terlihat pada Tabel
1.1.
Tabel 1 Tipe monazit dan rumus kimianya
Xenotim (YPO4) merupakan senyawa yttrium fosfat

yang mengandung 54-65% REE termasuk erbium, cerium
dan thorium. Xenotim juga mineral yang ditemukan dalam
pasir mineral berat, serta dalam pegmatit dan batuan beku.
Zirkon, merupakan senyawa zirkonium silikat yang
didalamnya dapat terkandung thorium, yttrium dan
cerium.
Gambar 8 Golongan Lanthanida dalam sistem periodik unsur
Mineral REE
REE tidak ditemukan di bumi sebagai unsur bebas
melainkan paduan berbentuk senyawa kompleks.
Sehingga untuk pemanfaatannya, REE harusdipisahkan
119
PROCEEDINGS
Gambar 10 Kenampakan mikroskopis Monazite (kiri) dan

Xenotime (kanan) daerah Bencah, Toboali, Bangka
REE tipe adsorpsi ion (Purawiardi, 2001). Cebakan primer

terutama berupa mineral bastnaesit, produksi terbesar
dunia dari China yang merupakan produk sampingan dari
tambang bijih besi. Cebakan yang lebih umum dikenal dan
diusahakan adalah cebakan sekunder, sebagian besar
berupa mineral monasit yang merupakan rombakan dari
batuan asalnya serta telah diendapkan kembali sebagai
endapan sungai, danau, delta, pantai, dan lepas pantai. Di
wilayah IUP PT Timah, cebakan REE yang sudah
teridentifikasi yaitu berupa tipe sekunder yang berasosiasi
dengan cebakan timah. Berdasarkan berat jenis nya,
mineral RE lebih ringan dibanding cassiterite sehingga
dalam suatu endapan aluvial, khususnya aluvial fan maka
naiknya kadar cassiterite selalu diikuti dengan
menurunnya MIT, begitu pula sebaliknya. Dengan
demikian, semestinya sebaran mineral RE akan lebih luas
dari cassiterite (Gambar 12).
Berdasarkan pendataan Litbangtek, stockpile monazite

di PPBT PT Timah sekitar 900 ton (400 ton berkadar
tinggi dan 500 ton berkadar rendah). Di pulau Bangka
sendiri, dilaporkan setidaknya terdapat lebih dari 8700 ton
stockpile monazite (6000 ton di area Kobatin, 2700 ton di
smelter swasta, Gambar 11).
Gambar 11 Stockpile monazite di Pulau Bangka (temuan

Litbangtek PT Timah, 2014)
TINJAUAN GEOLOGI MINERAL RE

REE tersebar luas dalam konsentrasi rendah (10 300
ppm) pada banyak formasi batuan. Kandungan unsur
tanah jarang yang tinggi lebih banyak dijumpai pada
batuan granitik dibandingkan dengan pada batuan basa.
Konsentrasi unsur tanah jarang tinggi dijumpai pada
batuan beku alkalin dan karbonatit. Berdasarkan
genesanya, cebakan mineral RE dibagi dalam dua tipe
(Tabel 2.1), yaitu cebakan primer sebagai hasil proses
magmatik dan hidrotermal, serta cebakan sekunder tipe
letakan sebagai hasil proses rombakan dan sedimentasi.
Pembentukan mineral RE primer dalam batuan karbonatit
menghasilkan mineral bastnaesit dan monazite
(http://minerals.usgs.gov). Batuan karbonatit sangat kaya
kandungan REE, dan merupakan batuan yang
mengandung REE paling banyak dibandingkan dengan
batuan beku (Verdiansyah, 2006).
Gambar 12 Ilustrasi Sebaran mineral pada tin aluvial deposit.
POTENSI THORIUM DI IUP PT TIMAH TBK.

PT. Timah, Tbk telah dalam rangka eksplorasi potensi
mineralisasi timah, telah melaksanakan rangkaian
kegiatan mulai dari tahapan pemetaan geologi (umum,
detail), geokimia, geofisika, pemboran (geologi, prospeksi
dan produksi), preparasi dan analisa sampel hingga
pemodelan perhitungan sumberdaya/cadangan timah dan
mineral ikutan yang terintegrasi dalam Satuan Kerja
Eksplorasi. Namun, dalam rangka memahami karakteristik
mineral ikutan timah pembawa unsur tanah jarang (UTJ)
khususnya monazite, penyusun membatasi data serta
metodologi penelitian sebagai berikut
Historical REE di PT Timah
Pada tahun 1973, Dinas Eksplorasi-Geologi melakukan
perhitungan (penaksiran) sumberdaya mineral ikutan
timah yaitu monazite, ilmenite, dan zircon, yang dihitung
berdasarkan nilai persentase mineral-mineral dalam satu
lembah. Adapun wilayah yang termasuk kedalam area
penaksiran yaitu Wilasi Pangkalpinang, Sungailiat,
Belinyu, Mentok, dan Toboali. Rumusan yang dipakai
untuk melakukan pendekatan perhitungan yaitu :
Tabel 2 Klasifikasi cebakan mineral RE (Industrial Mineral,

Special Review, 1991)
Dalam berbagai batuan, mineral RE pada umumnya

merupakan mineral ikutan (accessory minerals), bukan
sebagai mineral utama pembentuk batuan. Pada zonasi
pegmatit, unsur tanah jarang terdapat pada zona inti, yang
terdiri dari kuarsa dan mineral RE.Batuan Granit
pembawa oksida REE, Sn, W, Be, Nb, Ta, dan Th terdiri
dari Granit tipe S atau seri ilmenit. Iklim tropis yang panas
dan lembab menghasilkan pelapukan kimia yang kuat
pada granit. Pelapukan ini menyebabkan alterasi mineral
tertentu, seperti feldspar, yang berubah menjadi mineral
lempung. Mineral-mineral lempung seperti kaolinit,
montmorillonit dan illit, merupakan tempat kedudukan
Adapun hasil dari perhitungan tersebut didapatkan

sumberdaya seperti terlihat pada Tabel 3
Tabel 3. Estimasi Sumberdaya MIT
120
PROCEEDINGS
pemboran geologi, prospeksi dan produksi. Informasi

yang digunakan meliputi data collar (data posisi,
kedalaman pemboran bdll), dan assay (variasi litologi
secara vertikal, tebal lapisan dll) yang dikombinasikan
dengan data kadar hasil analisa laboratorium (kadar timah
dan mineral ikutan timah).
Untuk pemodelan/studi kasus digunakan data bor laut
yang berada di selatan Laut Ranggam IUP 1553 sebanyak
16 lubang bor yang mengandung MIT khususnya
monazite, dan data hasil bor bangka di daerah Pompong,
Bangka Selatan sebanyak 158 lubang bor Sedangkan
untuk perhitungan sumberdaya dengan menyeleksi
1.068.197 lubang bor, dan diperoleh 40.679lubang bor
untuk composite multimineral, termasuk bor khusus MIT
sebanyak 8.901.
Beberapa catatan yang dapat dijadikan acuan

diantaranya:
1) Angka cadangan MIT hendaknya selalu dianggap
sebagai taksiran belaka.
2) Grid pemboran cassiterite tidak bisa dipakai untuk
MIT, hanya ilmenite dan rutile yang masih dapat
menggunakan grid timah.
3) Naiknya kadar cassiterite selalu diikuti oleh
menurunnya MIT.
4) Untuk penanganan eksplorasi hanya diperlukan suatu
administrasi yang sederhana dan eksplorasi yang
dapat menjawab How Much (bagaimana mutu,
dsb.)
Kemudian di tahun 2012 saat REE mulai dinilai
sebagai prospek menjanjikan, dilakukan estimasi
sumberdaya REE di darat dan laut untuk kepentingan
inisiasi awal pilot project pengolahan ReOH. Adapun
dasar dari perhitungan kadar menggunakan data pemboran
MIT di Bangka (onshore) pada tahun 1999, sedangkan
volume nya didapatkan dari komposit data pemboran
tahun 2003-2009 (lihat Tabel 4).
Pemodelan Endapan
Sumberdaya Monazite
serta
Perhitungan
Potensi
Pemodelan endapan dikhususkan pada area insitu

dengan menyeleksi lubang bor yang belum menjadi tailing
serta memiliki kandungan MIT (monazite dll), kemudian
dibuat database untuk membuat profil vertikal litologi
yang disertai informasi kandungan cassiterite dan
monazite untuk selanjutnya dilakukan korelasi
berdasarkan urutan pengendapan litologi untuk
identifikasi hubungan litologi, endapan cassiterite dan
monazite.
Tabel 4 Estimasi Sumberdaya Hipotesis MIT, 2012
Gambar 13 Metoda Perhitungan Monazite dan unsur tanah

jarang
Perhitungan sumberdaya monazite dan unsur tanah

jarangnya,
dilakukan
sebagaimana
perhitungan
sumberdaya timah. Untuk mengetahui volume ore
monazite diperoleh melalui metoda block model dengan
memodelkan ore body kaksa dikalikan dengan kekayaan
kandungan monazite dan koefisien bor. Untuk kekayaan
unsur, yaitu dengan mengalikan kadar unsur rata-rata dari
hasil komposit bor tahun 2004 s.d 2013 dengan hasil
monazite dihitung.
Berikut ini adalah skema perhitungan sumberdaya
Thorium dan REE :
1) Mineral RE yang dihitung adalah Monazit, Xenotime,
dan Zircon. Sedangkan Unsurnya adalah cerium (Ce),
thorium (Th), dan yttrium (Y).
2) Volume ore didapatkan melalui metode block model
(micromine) dengan memodelkan lapisan kaksa (ore
body timah) pada semua wilayah IUP PT Timah
(Gambar 6).
Sumberdaya REE di IUP PT Timah

Sebagai langkah awal untuk memulai langkah strategis
dalam eksplorasi REE, maka dilakukan pendekatan
perhitungan sumberdaya REE pada seluruh wilayah IUP
PT Timah (Persero) Tbk. Tujuan dari perhitungan ini
untuk mendapatkan area-area anomali dengan kandungan
REE yang relatif tinggi baik dari segi kadar maupun
tonase nya.
Data
Data yang dimaksud adalah data pemboran yang
telah dilakukan sampai dengan 19 November 2015. Data
tersebut meliputi data bor dengan klasifikasi data
121
PROCEEDINGS
3)
4)
5)
Kadar RE mineral dihitung seperti perhitungan TDH,

yang data nya berdasarkan komposit data pemboran
per lembahtahun 2004-2013. Misalnya : Monazit
dihitung (MDH) akan didapatkan rumusan : MDH =
(Gr M / Tebal Ore) * KB
Kadar REE (unsur) berdasarkan perbandingan no
massa dalam tipe Monazite (Ce), yaitu Ce = 0.167, Y
= 0.106, dan Th = 0.277 (Tabel 5)
Tonase didapatkan dengan mengalikan Kadar dan
Volume.
Tabel 5 Jenis Monazite dan persentase thorium dalam Monazite
Gambar 14 Blok-blok Anomali REE Offshore

Tabel 7 Blok-blok Anomali REE Onshore.
Hasil Perhitungan
Berdasarkan hasil perhitungan, potensi terbesar REE
berada di wilayah Bangka, baik onshoremaupun offshore.
Tabel 6 di bawah ini merupakan rekapitulasi sumberdaya
seluruh wilayah IUP, untuk rincian masing-masing daerah
dapat dilihat pada lampiran. Tentu saja hasil perhitungan
lebih besar jika dibandingkan dengan perhitungan
terdahulu karena cakupan wilayahnya lebih menyeluruh
serta cara pendekatannya yang berbeda.
Tabel 6 Rekapitulasi Sumberdaya Monazite dan Thorium REE
di wilayah IUP PT Timah (Persero) Tbk.
Adapun area-area anomali dengan kandungan REE

yang relatif tinggi dapat dilihat pada Tabel 7 dan 8. Areaarea anomali tersebut memiliki kandungan kadar dan
tonase yang relatif tinggi jika dibandingkan dengan ratarata di wilayah sekitarnya, yang tentunya masih dalam
satu kondisi geologi yang sama. Daerah anomali dapat
dijadikan blok-blok prioritas untuk memulai kegiatan
eksplorasi REE di masa mendatang (Gambar 14, 15).
Pembagian blok-blok mengikuti administrasi wilayah
IUP.
Tabel 7 Blok-blok Anomali REE Offshore.
122
PROCEEDINGS
hubungan yang berbanding lurus antara cassiterite dan

monazite, baik besar butir maupun nilai kandungannya.
Besar butir cassiterite dan monazite dominan berada pada
ukuran relatif kasar, yaitu +48# s.d +65#. Demikian
persentase kandungannya yang sama-sama bernilai 1%
terhadap keseluruhan kandungan mineral pada sampel
yang ada. Kesebandingan tersebut diatas, kemungkinan
disebabkan adanya proses pemilahan besar butir pada saat
kegiatan pencucian timah pada saat penambangan
berlangsung, dengan berat jenis yang berdekatan, maka
butiran dengan fraksi yang sama akan terendapkan
bersamaan.
Gambar 15 Blok-blok Anomali REE Onshore di a) Pulau

Belitung dan b) Pulau Bangka
Gambar 17 Perbandingan Persentase Besar Butir dan Kandungan
Cassiterite dan Monazite
STUDI KASUS DAERAH POMPONG, BANGKA SELATAN

ONSHORE)
KESIMPULAN DAN TINDAK LANJUT

EKSPLORASI
Data bor pada daerah Pompong berada pada area

tailing, berbeda dengan studi kasus di awal yang berada
pada lingkungan insitu. Informasi dari hasil rekonstruksi
data bor, menunjukkan bahwa endapan monazite secara
dominan berada pada lapisan PKAKRK, LPPKA dan
LPPHA. Pada lapisan PKAKRK bahkan dominasi
monazite dengan kadar diatas > 0.1 kg/m mencapai
35%. Sedangkan pada lapisan LPPKA didominasi dengan
monazit berkadar 0.05 s.d 0.1 kg/m dan lapisan LPPHA
hampir merata. Endapan monazite tersebut muncul mulai
dari -4 s.d +18 m dari muka laut rata-rata, namun
konsentrasi yang relatif tinggi muncul pada -2 s.d +8 m.
Kesimpulan
Sumberdaya hipotetik mineral RE di wilayah IUP PT
Timah sekurang-kurangnya yaitu : 180.323 ton monazite,
21.876 ton xenotime, dan 1.226.268 ton zircon. Dengan
besar sumberdaya hipotetik Th orium 8.710 Ton
Perlu kajian lebih mendalam terkait REE di Bangka,
Belitung, Kundur serta tingkat keekonomisannya
berdasarkan sumberdaya yang ada. Sebagai catatan bahwa
dominasi REE di wilayah tersebut termasuk dalam
kategori light REE, seperti cerium (Ce) yang harganya
hanya sekitar $ 5000 diatas harga timah sehingga
sumberdaya nya pun harusnya hampir sebesar sumberdaya
timah.
Kebijakan eksplorasi tentang tindak lanjut REE harus
jelas, karena saat ini pemboran ketika sudah tidak ada
timah akan distop. Apakah diperlukan eksplorasi yang
khusus menangani REE ataukah hanya sebagai by
product saja. Hal ini akan menentukan rencana jangka
panjang terkait REE akan seperti apa.
Regulasi nasional tentang REE pun belumlah jelas,
sehingga PT Timah dalam hal ini dapat berperan sebagai
pihak yang support bagi pemerintah atau badan
berwenang dalam menentukan regulasi.
Tindak Lanjut Eksplorasi
Gambar 16 Litologi Yang Mengandung Endapan Monazite dan

Kandungan Monazite
Berikut ini merupakan langkah-langkah nyata dalam

jangka pendek yang akan dilakukan untuk menunjang
eksplorasi REE.
Hubungan cassiterite dan monazite pada tailing di

daerah Pompong berbeda dengan pada area insitu di Laut
Ranggam. Hasil rekonstruksi data memperlihatkan
a)
123
Geologi
Eksplorasi,
Melakukan
kegiatan
eksplorasi secara menyeluruh terhadap semua
PROCEEDINGS
Mineral Ikutan Timah (MIT). termasuk REE dan

dan Unsur Radioaktif (Th, U).
DAFTAR PUSTAKA
Chappel, B. W., dan White, A. J. R.. 2001. Two Contrasting
Granite Types: 25 Years Later. Australian Journal of Earth
Sciences. Hal. 489 499.
Cobbing, E. J., Pitfield, P. E. J., Darbyshire, D. P. F. and
Mallick, D. I. J. (1992). The granites of the South-East
Asian Tin Belt. British Geoloical Survey OverseasMemoir,
10: 369p.
Darmawan, B., 2012, Rare Earth Potential Indonesia in Tin
Belt, Presentasi pada FGD tentang REE, Badan Geologi,
Bandung.
Environmental
Magnetics
Business
Dept.
Mitsubishi
Corporation, 2014, Study Results of Technical Appraisal
for Extraction of REEs Based on the New Approach,
Presentasi di PT. Timah, Pangkalpinang, Bangka.
Hall, R., (2014). The origin of Sundaland. In Proceedings of
Sundaland Resources 2014 MGEI Annual Convention, 1718 November 2014, Palembang, South Sumatra, Indonesia:
p1-25.
Ishihara, S., (1981). The granitoid series and mineralization.
Economic Geology, 75th Anniversary Volume, 458-484.
Setijadji, L. D., Warmada, I. W., Imai, A., dan Sanematsu, K.,
2009, Investigation on Rare Earth Mineralisation in
Indonesia, dalam: Proceedings of The 2nd Regional
Conference Interdiciplinaries Research on Natural
Resources and Materials Engineering, Yogyakarta, p.
Soetopo, B., Subiantoro, L., dan Haryanto, D.,2012, Studi
Deposit Monasit dan Zirkon di Daerah Cerucuk Belitung,
Pusat Pengembangan Geologi Nuklir-BATAN, Jakarta
Selatan.
Sujitno, S., Rachmat, A., dan Harsono, Raf. 1973, Perspektif
Mineral-Mineral Ikutan dalam Lingkungan Timah dan
Mineral-Mineral Economic Lainnya di Bangka, Dinas
Eksplorasi-Geologi, P.N. Timah.
Suprapto, S.J., Tinjauan Tentang Unsur Tanah Jarang, Bidang
Program dan Kerja Sama Pusat Sumber Daya Geologi,
Bandung.
Suwargi, E., Pardianto, B., dan Ishlah, T., 2010, Potensi Logam
Tanah Jarang di Indonesia, Buletin Sumber Daya Geologi
Volume 5 Nomor 3 2010, Bandung.
b) Dilakukan perbaikan metoda preparasi conto, agar

kandungan MIT , REE dan unsur radio aktif dapat
tertangkap.
c)
Dilakukan validasi atas sumber daya setiap 3

bulan.
d) Laboratorium, Perubahan metode analisis di Lab

dengan lebih mengoptimalkan penggunaan XRF.
Pengadaan Protrace Elemen agar lebih akurat.
Update database tentang REE, misalnya :
kandungan dan komposisi unsur pada RE mineral
di Bangka, Belitung, dan Kundur.
e)
Pemboran, Pencucian sampel pemboran tidak

sampai dengan konsentrat (hanya sampai sekitar
10% dari sampel) untuk menghindari mineral RE
terbuang pada saat pendulangan. Baru dilakukan di
Geotin, untuk kapal mitra akan diberlakukan pada
kontrak tahun ini.
Target Persiapan Cadangan (stok bijih monazite saat

ini + 500 ton)
1) Tahun 2016 : 500 Ton Monazite (485 Ton REE + 15
Ton Thorium)
2) Tahun 2017 : 1,000 Ton Monazite (970 Ton REE +
30 Ton Thorium)
30 Ton Thorium)
30 Ton Thorium)
5) Tahun 2020 : 5,000 Ton Monazite (4,850 Ton REE +
150 Ton Thorium)
124
PROCEEDINGS
125
PROCEEDINGS
HVSR passive seismic surveying for mineral, coal or shallow quarry materials
exploration: three mineral exploration case studies.
David Stannard1*, Jayson Meyers1
1
34 Brisbane Street Perth 6000, Western Australia, Australia,

davids@respot.com.au, jaysonm@respot.com.au
ABSTRACT
The passive seismic method has been commonly applied to earthquake hazard mapping studies, engineering and structural
analysis applications, and sub-soil site characterisation investigations. An innovative seismometer developed by MoHo s.r.l
(Italy), the Tromino, has also proven itself as a rapid and inexpensive geophysical survey tool for assisting with regolith
and basin thickness mapping for exploration of mineral deposits, shallow sedimentary basin mapping and shallow quarry
materials by removing the requirement for large seismometer arrays, heavy and complicated instruments, and associated
long set up and recording times. In some instances, the Tromino assists by direct exploration by measurement of target
sediment thickness (e.g. ironsand), but typically indirect exploration by measuring low velocity and density regolith and
sedimentary cover thickness to help interpret and process other geoscientific datasets, such as gravity, electromagnetics
and for seismic static analysis, and to reduce drilling costs by providing infill information between holes and predicting
cover thickness ahead of drilling. The Tromino seismometer is a self-contained, lightweight survey instrument that is
highly portable and very simple to use.
The Tromino exploits the horizontal over vertical (H/V) spectral ratio (HVSR) technique, whereby the spectral ratio of the
averaged horizontal (H) and vertical (V) seismic velocity component data are used to determine the resonant frequency of
near surface layers over a strong basement acoustic impedance contact, which is also the seismic amplification frequency of
the sub-soil, at the recording location. This resonant frequency is related to the shear wave velocity and thickness of the
resonating layer by the formula: 0 = (4), where f0 is the fundamental frequency (in Hz) of the observed HVSR peak,
Vs is the shear wave velocity (in m/s) of the upper layer, and H (in m) is the depth to the acoustic impedance contrast. For
exploration purposes, the main acoustic impedance contrast of interest is commonly the regolith and sedimentary cover to
fresh rock interface or the base of a sedimentary basin, and lower modes are often related to impedance contrasts in the
sedimentary and regolith cover.
Resource Potentials Pty Ltd (Australia) are the Oceania representatives for geophysical equipment developed by MoHo,
and have utilised the Tromino for passive seismic surveying in Australia and abroad. The following paper briefly presents
the results of some passive seismic survey case studies using the Tromino for an REE oxide deposit and potash brine
project in Australia, and at an ironsand deposit in New Zealand. Additional case studies will be shown in the presentation.
acoustic impedance contrast boundaries in the sub-surface.
For mineral, coal and construction material exploration,
this strong acoustic impedance contrast is typically due to
low velocity and low density sediments overlying higher
velocity and denser bedrock.
MoHo s.r.l (Italy) addressed the restrictions of
traditional passive seismic arrays for widespread
exploration purposes, essentially total survey time, when
they developed the innovative Tromino seismometer
primarily for HVSR surveying. The Tromino is a highly
sensitive seismometer that addresses exploration industry
needs by incorporating all of the equipment required for
passive seismic array surveying within a small and highly
portable instrument that is easy to deploy in the field. The
Tromino ENGY 3G houses 3 velocimetric channels, 3
accelerometric channels, 1 analogue channel, inbuilt GPS,
inbuilt radio transmitter / receiver, internal A-D and data
recorder, and it is powered by 2x AA batteries that last all
day.
Resource Potentials Pty Ltd (Australia) are the Oceania
representatives for the innovative geophysical equipment
developed by MoHo s.r.l, including the Tromino
seismometer, and have used it to carry out passive seismic
survey case studies for exploration projects within
Australia, New Zealand and abroad. Three case studies
from passive seismic surveys carried out by Resource
Potentials using the Tromino instruments are briefly
presented in this paper to highlight some of the potential
INTRODUCTION
Passive seismic surveying is not yet commonly used as
a surveying method for mineral, coal or construction
material exploration, but the technique has been used for
decades for earthquake hazard mapping studies,
engineering
applications,
site
characterisation
investigations, and seismic mapping research studies. The
term passive seismic is given to the method, which is
really the recording of ambient seismic energy that exists
everywhere on the Earths surface, and is caused by
microseismic events, wind, waves, cultural sources, etc.
No external source, such as a hammer, weight drop,
explosives or vibroseis, is required to generate seismic
signal, and only a broadband receiver and survey location
is needed. Historically, the seismometers used for passive
seismic studies were typically set up in large complex
arrays (e.g. SPAC or FK), requiring several large
seismometers (>2) with substantial accompanying
equipment, long set up times, and very long recording
times (>4 hours per reading), and difficult data processing
and modelling approaches, thereby reducing the
attractiveness of the technique for exploration mapping
applications. The HVSR technique is a simple and
relatively rapid passive seismic method that provides
information on the resonant frequency and shear wave
velocity of near surface layers, and this information can be
used to provide a proxy for estimating the depth to strong
126
PROCEEDINGS
Fig. 9 Example of a fundamental seismic resonance frequency

peak at 2.33 Hz (top profile) generated by calculating the ratio
between the spectral component of the H and V components
(lower profile). Example data from a passive seismic survey
carried out in Esperance, Western Australia.
applications of this technique for exploration throughout

Indonesia and SE Asia.
PASSIVE SEISMIC HVSR DISCUSSION
Natural seismic energy occurs continuously within the
Earth due to microseismic tremors (microtremor), wind,
waves, anthropogenic sources, that typically produce high
energy surface waves (Rayleigh and Love waves). In
conventional seismic reflection surveys, this surficial
seismic energy is considered noise or ground roll and is
removed. However, for passive seismic surveying, this
natural or ambient energy is considered the signal.
In geological settings where a strong acoustic
impedance contrast occurs (e.g. unconsolidated sediments
overlying hard crystalline bedrock), vertically incident
SH-waves arriving from deep in the bedrock are trapped
within overlying lower velocity and lower density
sedimentary and regolith cover layers. This trapping of
energy, and associated constructive interference, generates
a fundamental resonance frequency within this cover
layer, which is also the seismic amplification frequency
that influence motion in overlying manmade structures,
such as buildings, bridges, dams, etc.
Using a seismometer to record 3 orthogonal
components of the natural seismic energy, two horizontal
components (Hx and Hy) and one vertical component (V),
the fundamental seismic resonance frequency of near
surface layers can be determined through the analysis of
the spectral ratio of the seismic components by using
recording times of 5 to 30 minutes, depending on the
depth of investigation and the nature of the ambient
signal. A Fast-Fourier-Transform (FFT) is used to convert
the 3 component time-series velocimetric data to
frequency spectra responses. The ratio between the
averaged H over the V component data will generate a
peak in the frequency at the sub-soil layer resonant
frequency. An example of this HVSR frequency peak is
shown in Figure 1.
The European SESAME project (Bard et al., 2004) and
many others (e.g. Nakamura 1989 and 2000; BonnefoyClaudet, 2008) have established that this fundamental
frequency, or the HVSR peak, is a good proxy to the SHwave resonance frequency (f0) which is related to the
shear wave velocity and thickness of the resonating layer
by a simple relationship:
Equation 1:
0 =
,
EXAMPLES IN MINERAL EXPLORATION

The following passive seismic surveying case studies
were carried out by, or under the guidance of Resource
Potentials (Australia) and the results are briefly presented.
A detailed discussion on the survey methodology and
results will be presented during the conference. All
surveys were acquired using 2x Tromino 3G ENGY
seismometers and were processed by Resource Potentials
using Grilla software. HVSR stations, as shown in Figure
1, were acquired along survey traverses and were used to
generate cross sections of HVSR responses. These HVSR
survey data were normalised in amplitude to the observed
maximum and minimum HVSR responses.
Rare earth elements (REE) oxide deposit, Western
Australia
The thickness of REE enriched weathered regolith
cover (e.g. saprolite and saprock) overlying fresh granite,
and colluvial cover deposits, was investigated using the
HVSR technique.
Surveying was carried out along E-W orientated
traverses spaced 500 m apart and stations were acquired
every 50 to 100 m (Figure 2). A recording time of 12
minutes was used. 18 recordings were acquired at drillhole
collar locations where fresh bedrock (granite) was
intersected. The observed HVSR fundamental frequency
was compared to the geologically logged depth to fresh
bedrock from drilling, and is shown in Figure 3 excellent
correlation occurs and the trend equation between the
peak HVSR frequency and depth to fresh bedrock was
determined and is associated with high correlation
coefficient. The equation from this borehole calibration
analysis was then used as one way to determine bedrock
depth from Tromino HVSR readings taken away from
boreholes.
where f0 is the frequency (in Hz) of the observed HVSR peak,

Vs is the shear wave velocity of the upper layer (in m/s) and H is
the depth to the acoustic impedance contrast (in m).
Fig. 3 Comparing the peak HVSR response (X axis) with the

depth to fresh bedrock as determined by drilling (Y axis) in the
project area. The relationship between the fundamental
frequency and depth to drilled bedrock is given by the equation,
and represented the trend line (in red).
The saprolite and fresh granite contact was imaged

clearly across each traverse at a typical depth of 50 m
(Figure 4). Subtle HVSR responses near the surface
correlated with the transported cover-saprolite interface as
127
PROCEEDINGS
responses in the near surface are associated with clay layers

defined in the GPR survey.
defined in the geological logs. The results of the passive

seismic survey have been used to identify drill target areas
to test for thicker zones of weathered material for
concentrations of REE. Furthermore, areas of shallow
bedrock are identified, and these areas can be considered
for locating heavy plant equipment, such as crushers and
ball mill grinding circuits, for mine site design and
planning.
Potash brines, Western Australia

HVSR surveying was carried out to map the lateral and
depth extents of paleochannel to assist exploration for
high grade potash brine solutions below a highly
conductive and difficult to access salt lake playa valley.
The paleochannel is filled with Cainozoic to Recent
porous sand and gravels, overlain by silts and clays, and
these sedimentary channel deposits are deposited on
Archaean granite and greenstone bedrocks, resulting in a
strong acoustic impedance contrast. 20 minute recordings
were acquired every 100 m along survey traverses,
orientated perpendicular to the expected paleochannel
path. The deep axis of the paleochannel was detected
below the salt lake valley floor and was estimated to be up
to 170 m deep (Figure 6). The HVSR survey results
successfully assisted planning expensive lake rig rotary
mud drill holes to target areas where significant volumes
of the brine solution occur. This same approach can be
used for mapping subsurface channel deposits to explore
for alluvial gold and gems, paleochannel uranium and
lithium deposits, shallow coal basins, groundwater
supplies, and general cover thickness mapping.
Fig.4: Normalised HVSR cross section and drilling information

overlain, highlighting the top of fresh bedrock HVSR response at
depth, and the base of transported cover with a weaker response
nearer the surface. Dashed white line shows the depth to fresh
rock by calibrating the HVSR peak response with the equation
from Equation 1. Dashed black line is the base of conductive
cover from other cover thickness investigations using helicopter
electromagnetic surveying and conductivity depth inversion.
Titanomagnetite ironsand deposit, New Zealand

Ground penetrating radar (GPR) surveying was carried
out at using a 35 MHz antenna configuration to investigate
the thickness of a titanomagnetite ironsand deposit on the
west coast of the North Island of New Zealand. GPR
signal penetration extended to a maximum depth of 20 m,
showing a great amount of detailed internal geometry of
the sand dune deposit, but was unsuccessful at
determining the base of the ironsand deposit throughout
much of the project area due to lack of penetration from
interlayered conductive clay bands and deep basement
beyond the limits of the GPR system. Passive seismic
surveying was carried out along GPR traverses to image
the strong acoustic contrast of the fresh basalt and
limestone contact at the base of the deposit. Each reading
used a recording time of 20 minutes.
A strong HVSR peak was observed at each station and
correlated with the geologically logged depth to basalt or
limestone, typically at 40 to 70 m depth. The GPR data
were overlain on the normalised HVSR section (Figure 5),
confirming that GPR signal penetration was not sufficient
to resolve the underlying basalt contact at the base of the
ironsand deposit. Subtle HVSR responses observed within
the ironsand deposit are associated with GPR reflectors
caused by narrow clay bands from volcanic ash layers
within the deposit.
Fig. 6 Plan image showing the deep axis of the paleochannel

within the Salt Lake valley. Insert shows assemblage of
normalised HVSR cross sections mapping the deep paleochannel
axis throughout the project area, and used to generate a 3D
surface of the base of the paleochannel and estimate the volume
of paleochannel sedimentary fill.
REFERENCES
Bard, P-Y., and SESAME participants, 2004, The SESAME
Project: An overview and main results, 13th World
Conference on Earthquake Engineering, Paper No. 2207.
Bonnefoy-Claudet, S., S. Baize, L. Fabian Bonilla, C. BergeThierry, C. Pasten, J. Campos, P. Volant, and R. Verdugo,
2008, Site effect evaluation in the basin of Santiago de
Chile using ambient noise measurements, Geophysics
Journal International, 1-13.
Nakamura, Y., 1989, A method for dynamic characteristics
estimation of subsurface using microtremor on the ground
surface. Q. Rep. RTRI, 30(1), 25-33.
Nakamura, Y., 2000, Clear identification of fundamental idea of
Nakamuras technique and its applications, Proc. 12WCEE,
No. 2656, 177-402.
ACKNOWLEDGMENTS
Fig. 5: Normalised HVSR cross section overlain with processed
GPR traverses and drilling information. The passive seismic
survey was able to detect the basalt unit that underlies the
ironsand deposit, and confirmed by drilling. Subtle HVSR
The authors would like to acknowledge the following

people who assisted in this study: Jrmy Magnon, Nigel
Cantwell, and Matthew Owers.
128
PROCEEDINGS
129

Proceeding MGEI 2016

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Proceeding MGEI 2016

Загружено:

Авторское право:

Доступные форматы

PROOCEDINGS OF UNCONVENTIONAL EXPLORATION TARGETS &

LATEST TECHNIQUES AND NEW TOOLS IN MINERAL AND COAL

5-6 October 2016, Bandung, West Java, Indonesia

Editor in Chief : Arif Zardi Dahlius

Paper Reviewer : Emmy Suparka

Cover Design by Dede Noor Iman Taufik

Arif Zardi Dahlius

35 Yi Sun1 and Jing Chen

Jing Chen, David R. Cooke, Jos Pique,

Krisztin Szentpteri, Tatzky Reza

51 Nur Wiwoho and Wahyu Sunyoto

61 Harman Setyadi, Budi Santoso, STJ

The metamorphic rock-hosted gold

Arifudin Idrus1*, Sukmandaru

Budhi Kumarawarman, Gde Handojo

Robby I. Rafianto, Henry A. Cahyono,

113 Yoseph C. A. Swamidharma

123 Setiawan Raharjo

The decline in Indonesias Exploration Performance-what can we do about it

Fig. 2 Size, type and location of significant mineral deposits in

Of these 10,334 deposits, 260 (or ~2.5%) are located

Fig. 1: Exploration expenditures and number of discoveries in

DATA AND METHOD

RESULT AND DISCUSSION

comparable to the World average rate of 45% over the

Fig. 3: Average cost per mineral discovery in Indonesia and

Figure 3 shows that, a similar, albeit less dramatic,

encourage local companies to take a more active role in

To do this the Government needs to re-engage with

Successful Techniques in Exploring the Lithocap Environment of the Sunda

RECENT MAJOR DISCOVERIES OF WORLD-CLASS

Moz Au and 8.6 Mt Cu (Intrepid Mines Ltd., 2012).

Hijau, the porphyry deposit is exposed at surface, with

LITHOCAP GEOMETRY AND EXPLORATION MODELS

SUCCESSFUL EXPLORATION TECHNIQUES LEADING TO

At regional and district scales and early exploration

2005) and Tumpangpitu (Harrison and Maryono, 2012;

technique to locate sizeable altered areas. One sample can

Porphyry and Epithermal Deposits: Example from the

lithocaps and understanding alteration zoning is crucial, as

Conference Proceedings: Gold Mining Journal and

Orogenic Gold Deposits - From Hypozonal to Epizonal Systems

Figure 1. Ages of Mineralization

DEFINITE ORE DEPOSIT CHARACTERISTICS

Pressure-temperature conditions of mineralization are

The review focuses on mineralization signatures

Table 1 indicates the host rock types of the investigated

Amphibolite; felsic schist

Shale, dolomite, gabbroic sills

North Limpopo thrust zone

Hutti-Maske shear zone

Synkinematic mafic intrusion

Mon Repos thrust zone

Main Uralian fault

Structurally epigenetic hypozonal and mesozonal

Table 2: Structural setting

Table 3: Timing of mineralization relative to metamorphism

Chitradurga boundary shear

A salient question for the formation of epigenetic

Table 1: Host rock lithology (HZ = hypozonal)

Cuiab tubular sheath fold

Consort Bar shear zone

syn- to late-peak metamorphism

Common assemblages associated