You are on page 1of 53

Indonesian Journal on Geoscience Vol. 1 No.

2 August 2014: 65-70

INDONESIAN JOURNAL ON GEOSCIENCE


Geological Agency
Ministry of Energy and Mineral Resources
Journal homepage: hp://ijog.bgl.esdm.go.id
ISSN 2355-9314 (Print), e-ISSN 2355-9306 (Online)

Seasonal variation of 13C content in Porites coral from Simeulue Island


waters for the period of 1993-2007
Sri Yudawati Cahyarini
Research Centre for Geotechnology LIPI, Kompleks LIPI, Jln. Sangkuriang, Bandung

Corresponding author: yuda@geotek.lipi.go.id


Manuscript received: March 27, 2014, revised: April 2, 2014, approved: July 14, 2014

IJ
O

Abstract - Variation of 13C content in coral skeletons shows the influence of metabolic fractionation in aragonite coral.
Understanding coral 13C variation can thus be useful to more understand e.g. past bleaching event which is further
useful for coral health and conservation. In this study, 13C content in Porites coral from Labuhan Bajau, Simeulue
Islands was analyzed. To know the correlation between variation of coral 13C and light intensity, the monthly variation of coral 13C is compared to solar radiation and cloud cover. The result shows that for the period of 2003 to 2008,
coral 13C shows it is well correlated (r=0.42 p=0.153) with cloud cover variation in annual mean scale. Meanwhile, in
seasonal mean variation, coral 13C is strongly influenced (r=0.85 p<0.0001) by cloud cover with 1 - 2 month time lag.
Comparing to the solar radiation (cloud cover), SST influences dominantly the variation of coral 13C from southern
Simeulue Island waters (LB sample) in an annual mean scale than in a seasonal scale.
Keywords: 13C, coral, Porites, solar radiation, cloud cover, SST

Introduction

A study of stable carbon isotopic composition


( C) content in coral skeletons shows that 13C
composition is primarily influenced by metabolic
fractionation (e.g. McConnaughey, 1989; Grottoli, 2002) i.e. influenced by photosynthesis and
respiration (McConnaughey,1989; Grottoli, 2002,
in Rodrigues and Grottoli, 2006). Decreased
coral 13C coinciding with bleached coral (Porter
et al.,1989, in Rodrigues and Grottoli, 2006) is
due to decreased photosynthesis, while increased
zooplankton decreased 13C is via respiration
(Felis et al.,1998; Grottoli and Wellington, 1999,
in Rodrigues and Grottoli, 2006).
Understanding 13C content in coral skeleton
is useful to more understand a physiological
factor which is a base to more understand the
bleaching event and further to the coral health
and conservation (Rodrigues and Grottoli, 2006).
In this study, 13C content in Porites coral from
13

Labuhan Bajo Simeulue waters was analyzed.


Monthly variation of coral 13C is compared to
the solar radiation and cloud cover to understand
the influence of light intensity to the coral 13C.
This study can be as a base for a historical coral
development study for a longer time window,
i.e. hundreds to thousands year ago. Further, the
study is able to support the coral reef conservation
activities in the Simeulue ocean waters.
Materials and Methods

Porites coral core was collected in July


2007 from 25 m depth of Labuhan Bajo Village, east-southern coast of Simeulue Island
(sample code LB) (~ 02.23o N 096.29o E). The
coral core was slabed into 0.5 cm thick and the
coral slabs were X-rayed with 3 Kvp to picture
clear density band. Water pressure cleaning
was done to the coral slabs to remove dust,

IJOG/JGI (Jurnal Geologi Indonesia) - Acredited by LIPI No. 547/AU2/P2MI-LIPI/06/2013, valid 21 June 2013 - 21 June 2016

65

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 65-70

then ultrasonic bath cleaning followed. X-rayed


coral slabs were then used to perform the coral
banding (Figure 1). Subsampling using hand
drilling (1 mm bit) was done along the coral
growth axis to get the coral powder samples.
Coral powder samples were then analyzed for
13C using Gasbench Delta Plus at the Free
University Amsterdam. The powdered samples
were reacted with H3PO4, and the resulting CO2
gas was analyzed in the mass spectrometer. All
samples are reported in .

Bottom

Top

Historical data used in this study involved


solar radiation, cloud cover, and coral SST. Solar
radiation data are obtained from Fresco v.6 averaged over 2x2 grid boxes resolution (from Wang
et al., 2008) and available from 2002- 2007.
Cloud cover data are obtained from ICOADS
with 2x 2 grid resolution and available from 19962007. Variable cldc (Cloudiness Monthly Mean
at Surface) is in okta. ICOADS data provided by
the NOAA/OAR/ESRL PSD, Boulder, Colorado,
USA, from their Web site at http://www.esrl.noaa.
gov/psd/.
SST derived from coral Sr/Ca (further mentioned as coral SST) was used in this study to
indicate the influence of SST to the variation of
coral 13C. Coral SST is obtained from Cahyarini,
(2011).

Figure 1. X -Rayed of Porites coral slab sample LB. Dashed


lines are subsampling transect along the growth axis.

Results and Discussions

The analysis of 13C content in coral skeleton


sample LB using Gasbench Delta Plus in monthly
resolution is shown in Figure 2. Monthly variation
of coral 13C (Figure 2) ranges from -3.390.42
to -2.070.42 with the mean value -2.980.42
for the period of 1993-2007. For the period of
2003 to 2008, decreasing trend of solar radiation
supposes decreasing trend of coral 13C from this
region (Figure 3), which confirms the published
work i.e. for the healthy coral as solar radiation
decrease, decreased coral 13C is due to decreasing

IJ
O

Coral XDS software was used to calculate the


paired high/low density which was used to developed preliminary chronology in annual scale.
One year growth is represented by a dark and light
coral band in x -rayed coral. The detailed chronology (i.e. monthly scale) of 13C is based on coral
Sr/Ca chronology development (see Cahyarini,
2011). Paired density band calculation result in
that LB Porites coral is about ~14 years old. The
chronology development of coral 13C results in
period ranges from July 1993 to August 2007.
Monthly variation of 13C from Porites coral
(sample code LB) for the period of 1993-2007 is
shown in Figure 2.
-2.0

-2.5

-3.0
-3.5

Figure 2. Monthly variation of 13C content in Porites coral (sample code LB) from Labuhan Bajo.

66

2007

2006

2005

2004

2003

2002

2000

1999

1998

1997

1996

1995

1994

1993

-4.5

2001

-4.0

900

0.00

800

-0.50

700

-1.00

600

-1.50

500

-2.00

400

-2.50

300

-3.00

200

-3.50

100
2002

2003

2004

2005

2006

2007

d13C

Solar radiation

Seasonal variation of 13C content in Porites coral from Simeulue Island waters for the period of 1993-2007
(S.Y. Cahyarini)

-4.00

Figure 3. Graphic of monthly variation of coral 13C (grey line) and solar radiation (dark line) and its trend lines (bold grey
and dark lines).

solar radiation coincides with low cloud cover


(Figure 4). In the studied areas, the maximum
solar radiation is in February (641.25 w/m2) and
the minimum is in August (495.09 w.m2), while
cloud cover maximum is in October (0.45) and
minimum is in Februari (0.27). Seasonal mean
variation of solar radiation and 13C coral shows
a good correlation in 1 month time lag. The maximum 13C is in November and the minimum is
in May. Figure 4 shows monthly mean variation
of solar radiation, cloud cover, and 13C coral.
Solar radiation supposes some time to reach the
coral in 25 depths to influence its 13C variation.

IJ
O

photosynthesis (Grottoli, 2002; Heikoop et al.,


2002). In some year periods, the seasonal cycles of
13C shows out of phase compared to solar radiation seasonal cycles, i.e. from 2005 to 2007, when
the high solar radiation coincides with low 13C.
This suggests that this period coincided with the
weak ENSO event when the seawater temperature
anomaly slowed down the enrichment of 13C in
the coral skeletons. During the normal condition,
coral 13C enrichment normally follows the solar
radiation cycle.
Seasonal mean variation of solar radiation
varies out of phase with the cloud cover, i.e. high

2.0

2.5

1.5

2.0

1.5

1.0

1.0

0.5

0.5

0.0

0.0

-0.5

-0.5

-1.0

-2.0

-2.0

-2.5

-2.5

Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec

-1.5

Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec

-1.0

-1.5

Figure 4. Graphic of monthly mean variation of solar radiation (grey line) and (left) cloud cover (dark line) and (right) 13C
(dark line). The data are corrected for two month lag. All time series data are standardized.

67

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 65-70

mean 13C of LB coral is correlated with cloud


cover (r=0.424 p=0.169) (Figure 6). The variation
of coral 13C relative to cloud cover variation is
about 0.123/okta (0 clear cloud to 8 overcast).
It suggests that the clearer the cloud, the more
13C content in coral.
Sr/Ca content in LB coral shows a local sea
surface temperature (SST) at a coral site (Cahyarini, 2011). Reconstructed SST based on Sr/Ca
content in LB coral (further mentioned as coral
SST) (Cahyarini, 2011) was used to understand
the influence of SST to the variation of coral
13C. Coral SST was compared and correlated
with coral 13C (Figure 7). The result shows that

Decreasing photosynthesis due to decreasing light condition may be caused by increasing


cloud cover, which causes decreasing 13C in coral
skeletons. Decreasing 13C content in Simeulue
coral is supposed to relate to decreasing light in
the depth of Porites coral (LB). This is convinced
that the maximum 13C amount in coral skeletons
occurred during a maximum light.
An annual mean 13C of LB coral sample and
cloud cover are compared (Figure 5). The result
shows that during ~10 years, from period of 1996
to 2007, decreasing trend of 13C coral follows
increasing trend of cloud cover. Correlation
between these two series convinces that annual
2
1.5
1
0.5

IJ
O

0
-0.5

-1

-1.5
-2

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Figure 5. Annual mean variation of 13C (grey line) and cloud cover (dark line). Linear trend line (dashed line). Data are
standardized to unit variance.

-2.0

y = -0.123x - 2.254
R2 = 0.180R=0.424

-2.2

-2.4

-2.6
-2.8

-3.0
-3.2
-3.4
3.5

4.0

4.5

5.0
5.5
Cloud Cover

Figure 6. Linear regression of cloud cover and 13C in the annual mean scale.

68

6.0

6.5

7.0

Seasonal variation of 13C content in Porites coral from Simeulue Island waters for the period of 1993-2007
(S.Y. Cahyarini)

2
1.5
1
0.5
0
-0.5
-1
-1.5
-2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Figure 7. Annual mean variation of coral 13C (dark line) and SST (grey line) derived from coral Sr/Ca. Data are standardized to unit variance.

References

Cahyarini, S.Y., 2011. Rekonstruksi suhu permukaan laut periode 1993-2007 berdasarkan
analisis kandungan Sr/Ca koral dari wilayah
Labuan Bajo, Pulau Simeulue. Jurnal Geologi
Indonesia, 6 (3), p.129-134.
Felis, T., Patzold, J., Loya, Y., and Wefer, G.,
1998. Vertical water mass mixing and plankton blooms recorded in skeletal stable carbon
isotopes of a Red Sea coral. Journal of Geophysical Research,103 (30), p.730-731.
Grottoli, A.G. and Wellington, G.M., 1999. Effect of life and zooplankton on skeletal d13C
values in the eastern Pacific corals Pavona
clavus and Pavona gigantean. Coral Reefs,18,
p.29-41.
Grottoli, A.G., 2002. Effect of life and bring
shrimp on skeletal d13C in the Hawaiian coral
Porites compressa: a tank experiment. Geochimica et Cosmochimica Acta, 66, p.1955-1967.
Heikoop, J.M., Hickmott, D.D., Risk, M.J.,
Shearer,C.K., and Atudorei,V., 2002. Potential
climate signals from the deep sea gorgonian
coral Primnoa resedaeformis. Hydrobiologia,
471, p.117-124.
McConnaughey, T., 1989. 13C and 18O isotopic
disequilibrium in biological carbonates: II
In Vitro simulation of kinetic isotope effect.
Geochimica et Cosmochimica Acta, 53,
p.163-171.

IJ
O

in annual mean resolution, correlation between


coral SST and 13C is high (r=0.54 p=0.057). It
suggests that the influence of SST to the variation of 13C is higher than that with cloud cover
(solar radiation) in annual mean scale. In seasonal
variation, coral 13C change respond to SST is low
(r=0.387 p<0.0001). It supposes the SST variation is dominant in changing the 13C content in
LB coral in the annual mean scale rather than in
seasonal scale.

Conclusions

The 13C variation content in coral from Labuhan


Bajo (LB) east southern Simeulue Island is well
correlated with cloud cover variation in annual
mean scale. In seasonal scale, variation of 13C
content in LB coral is influenced by cloud cover
with 1-2 month time lag. Variation of coral 13C
is dominantly influenced by SST in annual mean
than in seasonal variation.

Acknowledgements
The author acknowledges the KNAW Mobility
Program 2007/08 -grant to SYC for the isotope
analysis at the Isotope laboratorium facilities at
Vrije University, Amsterdam.

69

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 65-70

Hawaiian corals. Geochimica et Cosmochimica Acta, 70, p.2781-2789, doi 10.1016/j.


gca.2006.02.014.
Wang, P., Stammes, R. van der A.P., Pinardi, G.,
and Roozendael, M. van, 2008. FRESCO+:
an improved O 2 A-band cloud retrieval
algorithm for tropospheric trace gas retrievals. Atmospheric Chemistry and Physics, 8,
p.6565-6576.

IJ
O

Porter J.W., Fitt, W.K., Spero, H.J., Rogers,


C.S., and White, M.W., 1989. Bleaching in
reef corals: physiological and stable isotopic
responses. Proceedings of Natural Academic
Science,USA, 86, p.9342-9346.
Rodrigues, L.J. and Grottoli, A.G., 2006. Calcification rate and the stable carbon, oxygen, and
nitrogen isotopes in the skeleton, host tissue,
and zooxanthellae of bleached and recovering

70

Indonesian Journal on Geoscience Vol. 1 No. 2 August 2014: 71-81

INDONESIAN JOURNAL ON GEOSCIENCE


Geological Agency
Ministry of Energy and Mineral Resources
Journal homepage: hp://ijog.bgl.esdm.go.id
ISSN 2355-9314 (Print), e-ISSN 2355-9306 (Online)

Epithermal Gold-Silver Deposits in Western Java, Indonesia: Gold-Silver


Selenide-Telluride Mineralization
Euis Tintin Yuningsih1,2, Hiroharu Matsueda2, and Mega Fatimah Rosana1
Faculty of Geology, Padjadjaran University, Jln. Raya Bandung - Sumedang Km. 21, Jatinangor, Indonesia
2
The Hokkaido University Museum, Hokkaido University, Japan

Corresponding author: etintiny@yahoo.com


Manuscript received: January 17, 2014, revised: June 24, 2014, approved: July 19, 2014

IJ
O

Abstract - The gold-silver ores of western Java reflect a major metallogenic event during the Miocene-Pliocene
and Pliocene ages. Mineralogically, the deposits can be divided into two types i.e. Se- and Te-type deposits with
some different characteristic features. The objective of the present research is to summarize the mineralogical and
geochemical characteristics of Se- and Te-type epithermal mineralization in western Java. Ore and alteration mineral
assemblage, fluid inclusions, and radiogenic isotope studies were undertaken in some deposits in western Java combined
with literature studies from previous authors. Ore mineralogy of some deposits from western Java such as Pongkor,
Cibaliung, Cikidang, Cisungsang, Cirotan, Arinem, and Cineam shows slightly different characteristics as those are
divided into Se- and Te-types deposits. The ore mineralogy of the westernmost of west Java region such as Pongkor,
Cibaliung, Cikidang, Cisungsang, and Cirotan is characterized by the dominance of silver-arsenic-antimony sulfosalt
with silver selenides and rarely tellurides over the argentite, while to the eastern part of West Java such as Arinem
and Cineam deposits are dominated by silver-gold tellurides. The average formation temperatures measured from
fluid inclusions of quartz associated with ore are in the range of 170 220C with average salinity of less than 1 wt%
NaClequiv. for Se-type and 190 270C with average salinity of ~2 wt% NaClequiv. for Te-type.
Keywords: epithermal gold-silver deposit, fluid inclusions, selenides, Se-type, tellurides, Te-type, western Java

Introduction

Western Java hosts several gold deposits and


all of the mineralizations follows the Sunda-Banda magmatic arc, which is the longest magmatic
arc in Indonesia (Figure 1). The ore deposits of
western Java reflect a major metallogenic event
during the Miocene-Pliocene. Mineralogically,
the deposits can be divided into two types, those
are Se-type and Te-type with some different
characteristic features. Telluride and selenide
minerals in many epi- and mesothermal deposits
are often associated with gold and silver that
have an important role worldwide. The principal
characteristics of the Te- and Se- minerals in
epithermal deposit were described by Sillitoe and
Hedenquist (2003).

The Se-type of western Java mineralization


mostly lies within and on the flanks of the Bayah
Dome and is represented by Pongkor, Cikidang,
Cisungsang, Cirotan, and Cibaliung deposits,
while the Te-type is located more eastern and
represented by Arinem and Cineam deposits
(Figure 2). Studies of ore mineralogy and geochemistry were carried out within the epithermal
ore deposits of western Java by previous authors
such as Pongkor (Basuki et al., 1994; Marcoux
and Milesi, 1994; Sukarna et al., 1994; Milesi et
al., 1999; Sukarna, 1999; Warmada et al., 2003;
Syafrizal et al., 2005; Syafrizal et al., 2007;
Warmada et al., 2007), Cikidang (Rosana and
Matsueda, 2002), Cibaliung (Sudana and Santosa,
1992; Marcoux and Milesi, 1994; Marjoribanks,
2000; Angeles et al., 2002; Harijoko et al., 2004;

IJOG/JGI (Jurnal Geologi Indonesia) - Acredited by LIPI No. 547/AU2/P2MI-LIPI/06/2013, valid 21 June 2013 - 21 June 2016

71

100 E

125 E

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 71-81

Philippines

Malaysia

Central Kalimantan arc


Malaysia
Kalimantan

Sumatra

500

Sulawesi
East Mindano arc

Halmahera arc

Sulawesi

Irian Jaya

Sumatra-Meratus arc

Java

1,000

kilometers

P
N
G

Medial
Irian Jaya arc

Sunda-Banda arc

Australia

125 E

Late Miocene and Pliocene


Paleogene and Mid Tertiary
Late Cretaceous

Magmatic arc

IJ
O

Figure 1. Distribution of the magmatic arc within the Indonesia archipelago from Late Cretaceous to Pliocene (modified
after Carlile and Mitchell, 1994). The location of the studied area is bounded by a rectangle.

Serang

25 km

JAKARTA

Pandeglang

Cibaliung

Cikidang
Cirotan

Pongkor Bogor
Cisungsang

Sukabumi

7 S

Pelabuhanratu

Bandung

Cineam

Indian Ocean

106 E

Arinem

107 o E

Figure 2. Location and distribution of the Se- () and Te- () type epithermal Au-Ag deposits in the western Java, Indonesia.

Harijoko et al., 2007), Cisungsang (Rosana et al.,


2006), Cirotan (Milesi et al., 1993; Marcoux et
al., 1993), Arinem (Yuningsih et al., 2012), and
Cineam (Widi and Matsueda, 1998).
Most of the Se- and Te-type deposits in western Java are in the form of vein. However, the
Cisungsang deposit forms the massive sulfide
72

with some vein association. Vein size of the Seand Te-types are various from several meters
to more than 5,000 m in length and from a few
centimeters up to 5 m in width. The gold mineralization ages within this area for the Se-type are
mostly of Pliocene and Pleistocene with the range
from 2.4 to 1.7 Ma and Late Miocene (11.18 Ma)

Epithermal Gold-Silver Deposits in Western Java, Indonesia: Gold-Silver Selenide-Telluride Mineralization


(E.T. Yuningsih et al.)

for Cibaliung deposit. K-Ar age dating of Te-type


indicates the mineralization ages are around 9.9
~ 8.5 Ma or Late Miocene, respectively.
The principal objective of this research is to
summarize the mineralogical and geochemical
characteristics of Se- and Te-type epithermal
mineralization in western Java. Ore and alteration mineral assemblage, fluid inclusions, and
radiogenic isotope studies were undertaken in
some deposits in western Java combined with
literature study from previous authors.

Methods

Accuracy between -60 and -10C is estimated


in the order of 0.2C, whereas between -10
and +30C and above +200C is placed at 0.5
and 2C, respectively. Instrumental calibration was done using synthetic pure H2O (0C),
dodecamethylene Glycol (82.0C), benzanilide
(163.0C), sodium nitrate (306.8C), n-tridecane
(-5.5C), n-dodecane (-9.6C), chlorobezene?
(-45.6C), and chloroform (-63.4C) inclusion
standards.
Salinity was determined from the last melting temperatures of ice, utilizing the equation by
Bodnar (1993). The possibility of the presence of
volatile species (CO2, N2), hydrocarbons (CH4,
C2H6), and solid phases in fluid inclusions was
identified by Raman spectroscopic analyses on
limited samples.
Results and Analyses

Ore Mineralogy
The dominant opaque minerals from the Setype deposits are Se- and Se-bearing silver minerals (aguilarite, naumannite, argentite, polybasite,
and pyrargyrite), electrum, and tetrahedrite with
various amounts of sulfide minerals of sphalerite,
galena, chalcopyrite, arsenopyrite, and pyrite.
Other ore minerals are found in a trace amount.
Some rare minerals of Bi- and Sn-bearing minerals such as lillianite and canfieldite occur in
Se-type deposit of Cirotan (Milesi et al., 1993).
The Te-type is characterized by the occurrence
of hessite, petzite, stutzite, tetradymite, altaite,
and tennantite-tetrahedrite, with a high amount
of sulfide minerals of sphalerite, galena, chalcopyrite, and pyrite with occurrences of arsenopyrite. Some photomicrographs of the ore minerals
associated in the Se- and Te-type deposits are
presented in Figure 3.
Rare telluride minerals of hessite and altaite
were reported from the Se-type deposit (Harijoko
et al., 2007), but until now there are no selenide
minerals observed in the Te-type deposits of
Arinem, except for the Te-type deposit of Cineam
which contains trace of Se-bearing minerals of
pyrargyrite-proustite. The occurrences of the ore
minerals from the two types of deposits are sum-

IJ
O

Thin-, polished- and doubly polished sections


of samples from western Java deposits were
analyzed using transmitted- and reflected-light
microscopes. Additional samples of altered host
rocks were investigated by X-ray diffraction using standard treatment methods for clay mineral
identification. Geochemical analyses for major,
minor, and trace elements of ores were conducted
by ICP at Acme Analytical Laboratories (Vancouver) Ltd., British Columbia, Canada.
The compositions of ore minerals were determined using a JEOL 733 electron microprobe
analyzer at Hokkaido University. Standards used
were natural chalcopyrite, InP, MnS, CdS, FeAsS,
Sb2S3, PbS, SnS, HgS, ZnS, and elemental Se, Au,
Ag, Te. The probe was operating at 20kV voltage
and the beam current of 10nA was focused to 1-10
m diameters with peak counting for 20s. The
X-ray lines measured were As, Se, Te, Cd, Ag,
Bi, and Sb (L), S, Cu, Zn, Fe, and Mn (K), and
Pb, Au, and Hg (M). The data were corrected
by ZAF correction.
Doubly polished thin sections were prepared
on 200 m thickness for fluid inclusion study
on quartz, sphalerite, and calcite minerals. Microthermometric analysis was performed on a
Linkam THMSG 600 system attached to a Nikon
transmitted-light microscope. Heating rate was
maintained near 2C min-1 for measurement of
homogenization temperature (Thtotal) and 0.5C
min-1 for measurement of ice melting temperature
(Tm). Precision was calculated as 0.1C in the
temperature range of the observed phase changes.

73

IJ
O

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 71-81

Figure 3. Reflected-light photomicrographs of the ore mineral association from some deposits at the western Java. (a)
Pongkor; (b) Arinem; (c) Cikidang; (d) Cisungsang. Abv.: alt=altaite, arg=argentite, canf=canfieldite, cpy=chalcopyrite,
elm=electrum, gn=galena, hs=hessite, lim=limonite, pr=proustite, py=pyrite, pyrg=pyrargyrite, qtz=quartz, sph=sphalerite.

marized in Table 1 along with other characteristics


of those deposits.

Ore Geochemistry
The FeS content of sphalerite from the Tetype is generally similar to those of the Se-type
mostly in the range of 0.1-2.4 mol% (Se-type)
and 0.5~2.0 mol% (Te-type, rare are up to 8.5
mol%). However, the FeS content of sphalerite
from massive deposit of the Cisungsang (Se-type)
is higher, ranging from 13.6-19.6 mol%, and from
Cirotan is between 0.5 and 26.0 mol% (Milesi
et al., 1993). Cadmium content in sphalerite of
Se-type is in the range of 0.1-2.0 mol% and in
Te-type of Arinem around 0.1-1.0 mol%.
The Ag content of electrum from the Se-type
is higher than that from the Te-type, ranging
between 22-68 wt% and 14-40 wt%. Some ore

74

minerals from Se-type contain selenium such as


in galena which is up to 1.5 wt%, in acanthiteaguilarite up to 13.5 wt%, and in polybasite up
to 3.6 wt% (with Te content up to 5.5 wt%).
Tellurium content in proustite is in trace amount
and in uytenbogaardtite is up to 0.8 wt%. Ore
minerals of the Te-type deposit of Arinem contain
selenium such as in galena which is up to 1.9
wt%, in tetradymite 0.1-2.1 wt%, and up to 1.4
wt% in petzite.
Geochemical analyses on the bulk vein samples inferred Mn are higher in the Se-type, but
low in the Te-type. Bi and Hg are lower in the
Se-type compared to the Te-type deposit. The
comparison of the geochemical composition
between the Te- and Se-type deposits represented
by the Arinem and Pongkor deposits is show in
the Table 2.

IJ
O

Se- & Te-minerals: aguilarite-naumannite, trace altaite, hessite.


Sulfosalt: tennantite-tetrahedrite, POLYBASITE.
Sulfide & other minerals: pyrite, sphalerite, galena,
arsenopyrite, marcasite, ARGENTITE-stromeyerite, trace
chalcopyrite, bornite, electrum, rare native silver.

Se- & Te-minerals: aguilarite.


Sulfosalt: PYRARGYRITE, PROUSTITE.
Sulfide & other minerals: sphalerite, galena, arsenopyrite,
pyrite, marcasite, pyrrhotite, chalcopyrite, ARGENTITE,
greenockite, acanthite, canfieldite, electrum.

Se- & Te-minerals: aguilarite. Sulfosalt: tetrahedrite,


POLYBASITE, PYRARGYRITE. Sulfide & other
minerals: pyrite, marcasite, galena, chalcopyrite, pyrrhotite,
sphalerite, arsenopyrite, acanthite, greenockite, mackinawite,
uytenbogaardtite, covellite, electrum, scheelite, cassiterite,
canfieldite, lillianite, wolframite.

Cibaliung 3

Cisungsang4

Cirotan 5

Host rocks: calc-alkaline rhyolitic-dasitic, quartz microdiorite (Miocene) cutting


volcano-sedimentary series.
Gangue & alteration: chlorite, epidote, quartz, calcite, rare Fe-Ti oxides, illite/
smectite, kaolinite, adularia, apatite, rare gypsum, anhydrite.
Mineralization age: 1.7

Host rocks: breccia tuff, limestone (Miocene).


Gangue & alteration: not well developed, silica, small calcite, rare clay.
Mineralization age: -

Host rocks: basaltic andesite volcanics intercalated tuffaceous sediment (MiddleLate Miocene) overlaid unconformably by dacitic tuff, younger sediments & basalt
flows. The sedimentary rocks consist of conglomerate, calcareous sandstone, and
limestone.
Gangue & alteration: quartz, adularia, calcite, smectite, illite, mixed layered
chlorite-smectite and illite-smectite, kaolinite, epidote, zeolite.
Mineralization age: 11.18~10.65

Host rocks: volcanic rock of lapilli tuff & breccia (Early Miocene).
Gangue & alteration: smectite/chlorite mixed layer mineral, epidote, carbonate,
illite, quartz, kaolinite, limonite, montmorilonite, K-feldspar, adularia.
Mineralization age: 2.4

Host rocks: andesitic breccia, tuff, lapili, andesite intercalated limestone & sandstone
(Oligocene-Early Miocene), basement shale and sandstone, host rock overlain by
sedimentary rocks of Miocene age, which consist of claystone, limestone, sandstone,
and volcanic rocks.
Gangue & alteration: quartz, calcite, dolomite, kutnohorite, carbonate,
rhodochrosite, adularia, montmorilonite, chlorite, illite/smectite, smectite, kaolinite,
K-feldspar.
Mineralization age: 2.05~2.7

Host rocks, gangue, and alterationmineralization age (Ma)

pH: neutral.
Th: quartz: 180-255
sphalerite: 207-280
Salinity: sphalerite : 2.89-7.15
quartz: -

pH: neutral.
Th: quartz: 160-300
Salinity: quartz: 2.2-3.4

pH: neutral.
Th: quartz: 160-330
(170-shallow),
170-300 (220-deep)
Salinity: quartz: <1.0

pH: neutral.
Th: quartz: 170-260 (215)
Salinity: quartz: <3.0

pH: neutral.
Th: carbonate: 171-249
(20515)
quartz: 180-287 (22021)
spha : 220-320 (25829)
Salinity: carbonate: (0.50.6)
quartz: 0-5.0 (1.01.0)

pH, Th (C),
Salinity (wt% NaClequiv.)

Words in capital: Se- and or Te-bearing mineral. References: 1Basuki et al., 1994; Marcoux and Milesi, 1994; Milesi et al., 1999; Sukarna et al., 1994; Sukarna, 1999; Syafrizal et al., 2005; Syafrizal et al.,
2007; Warmada et al., 2003; Warmada et al., 2007; 2 Rosana and Matsueda, 2002; 3Angeles et al., 2002; Harijoko et al., 2004; Harijoko et al., 2007; Marcoux and Milesi, 1994; Marjoribanks, 2000; Sudana and
Santosa, 1992; 4Rosana et al., 2006; 5Milesi et al., 1993; Marcoux et al., 1993; 6Yuningsih et al., 2012; 7Widi and Matsueda, 1998.

Se- & Te-minerals: aguilarite.


Sulfide & other minerals: pyrite, ARGENTITE, sphalerite,
galena, electrum, manganese oxide, limonite.

Se- & Te-minerals: aguilarite, trace hessite.


Sulfosalt: tetrahedrite, famatinite, PROUSTITE, PEARCEITEPOLYBASITE, antimony-PEARCEITE. Sulfide &
other minerals: pyrite, chalcopyrite, sphalerite, galena,
uytenbogaardtite, chalcocite, stromeyerite, mckinstryite,
ACANTHITE, bornite, electrum, Au-Ag alloy,willemite,
massicot, manganese oxide, limonite.

Ore mineralogy

Cikidang 2

Se-Type
Pongkor 1

Deposit

Table 1. Summary of Mineralogic, Age and Geochemical Characteristics of the Se- and Te-Types of Western Java Ore Deposits
Epithermal Gold-Silver Deposits in Western Java, Indonesia: Gold-Silver Selenide-Telluride Mineralization
(E.T. Yuningsih et al.)

75

76

Se- & Te-minerals: hessite, petzite, aguilarite.


Sulfosalt: tetrahedrite-tennantite, PYRARGYRITE,
PROUSTITE.
Sulfide & other minerals: pyrite, sphalerite, galena,
arsenopyrite, chalcopyrite, argentite, realgar, stibnite,
orpiment, electrum, iron-oxide.

Se- & Te-minerals: hessite, altaite, tetradymite, stutzite,


petzite.
Sulfosalt: enargite, tennantite, tetrahedrite.
Sulfide & other minerals: sphalerite, GALENA, chalcopyrite,
pyrite, marcasite, arsenopyrite, pyrrhotite, argentite, covellite,
chalcocite, electrum, hematite.

Ore mineralogy

Host rocks: andesitic-dasitic volcanic rocks (Oligocene-Miocene) intruded by


diorite, granodiorite, andesite and dacite intrusive.
Gangue & alteration: quartz, illite, calcite, propylite, argillic, silisification and
locally phyrophylite.
Mineralization age: 8.5~9.6

Host rocks: andesitic tuff, tuff breccias, lava (Oligocene-Middle Miocene) overlain
unconformably by a volcanic group composed of andesitic breccias and tuff.
Gangue & alteration: quartz, calcite, sericite, illite, mixed layered illite-chlorite,
chlorite-smectite, kaolinite.
Mineralization age: 8.8~9.9

Host rocks, gangue, and alterationmineralization age (Ma)

IJ
O
pH: neutral.
Th: quartz: 190-240, up to 350
Salinity: quartz: 1.45-2.30,
up to 3.7.

pH: neutral.
Th: quartz: 157-327 (194-267)
sphalerite: 153-275 (194235)
calcite: 140-217 (187)
Salinity: quartz: 0.2-4.3
(1.6-2.7)
sphalerite: 0.9-3.9
(1.7-2.7)
calcite: 1.2-3.9 (2.6)

pH, Th (C),
Salinity (wt% NaClequiv.)

Words in capital: Se- and or Te-bearing mineral. References: 1Basuki et al., 1994; Marcoux and Milesi, 1994; Milesi et al., 1999; Sukarna et al., 1994; Sukarna, 1999; Syafrizal et al., 2005; Syafrizal et al.,
2007; Warmada et al., 2003; Warmada et al., 2007; 2 Rosana and Matsueda, 2002; 3Angeles et al., 2002; Harijoko et al., 2004; Harijoko et al., 2007; Marcoux and Milesi, 1994; Marjoribanks, 2000; Sudana and
Santosa, 1992; 4Rosana et al., 2006; 5Milesi et al., 1993; Marcoux et al., 1993; 6Yuningsih et al., 2012; 7Widi and Matsueda, 1998.

Cineam 7

Te-Type
Arinem 6

Deposit

Table 1. ......................continued

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 71-81

Epithermal Gold-Silver Deposits in Western Java, Indonesia: Gold-Silver Selenide-Telluride Mineralization


(E.T. Yuningsih et al.)
Table 2. Comparison of Bulk Chemical Analyses of Te-Type (represented by Arinem Deposit) and Se-Type (represented
by Pongkor Deposit) Ores
Arinem Deposit

Se
Te
Pb**
Mn
Cd
As
Sb
Hg
Bi

Arinem Vein
IIA
IIA

IA

IB

IIA

IIB

IIC

308
32
2.85
929
1,287.3
13.2
4.5
2.52
11.6

30
na*
0.31
696
47.2
>10,000
71.2
0.88
0.4

116
156
0.03
1,161
2.9
13.6
1.2
0.23
199.5

305
57
3.66
3,407
112.1
289.9
8.0
3.81
12.7

190
53
2.35
1,161
528.7
247.8
6.4
4.4
18.3

CGH

CGH

CGH

CGH

CRG

CRG

GH

14.5
<0.1
8
25,553
<0.1
10.2
12.4
45
0.06

12.0
0.5
10
2,729
<0.1
39.2
12.6
93
0.05

<0.1
<0.1
18
963
0.1
98.3
14.7
214
0.14

8.28
2.7
2,010
1,194
3.9
62.6
71.8
158
0.11

23.6
<0.1
770
1,039
32.1
22.9
34.1
18
0.09

52.0
<0.1
502
1,738
10.7
36.6
8
53
0.09

777
0.9
9,200
11,448
107
185
39.1
95
0.14

IIC

35
119
>500
14
14
209
1.24
1.32
15.71
1,703
309
2,013
101.7
885.3 >2,000
3,134.9 1,669.8
4.4
26.0
16.3
11.9
3.32
1.59
4.5
0.2
2.4
24.7

Bantarhuni Vein
IIB
IIC
IIC
212
na
0.10
>100
19.2
<0.5
2.8
0.06
122.1

>500
na
25.27
1,084
1,964.1
<0.5
24.3
3.22
42.9

>500
na
21.78
852
1,160.2
<0.5
15.4
1.44
6.3

KC

KC

KC

KC

181
3.1
10,100
5,248
101
58.9
125
123
0.14

199
0.2
438
259
5.3
46.1
146
19
0.1

145
0.2
276
747
3.2
53.7
156
69
0.09

44.3
<0.1
37
1,421
0.5
46.3
83.8
38
0.08

IJ
O

Se
Te
Pb
Mn
Cd
As
Sb
Hg***
Bi

Pongkor Deposit 1

Unit in ppm; * no analyses; ** in percent (%); *** in ppb; 1 Warmada et al. (2003).

Host Rocks and Hydrothermal Alteration


In general, the dominant host rocks for both Seand Te-type deposit are dacitic-basaltic volcanic
rocks. The Se-type occurs in volcanic rocks and
sometimes some sedimentary rocks intercalated;
while Arinem and Cineam deposits of the Te-type
formed in volcanic rocks. The gangue mineral of
Se- and Te-types is characterized by the presence
of large amounts of quartz, followed by carbonate
and illite. Some of Mn carbonate (manganoan calcite and rhodocrosite) occurred in Se-type deposit.
The characterizing gangue by the presence
of adularia is well developed in some Se-type,
while there is no adularia found at the Te-type.
Hydrothermal alteration patterns in the Se- and
Te-type deposits are similar with the abundance
of the propylitic alteration, and are dominated
by chlorite, illite, mixed layered illite-smectite
and chlorite-smectite. Argillic alteration is characterized by illite, montmorillonite, and some of
kaolinite, usually enveloping the vein where the
silicification and sericitisation occurred. The
association alteration minerals in those deposits

indicate the pH is neutral with slightly acid at the


late stage of mineralization for some of deposit
(such as in Arinem).
Homogenization Temperature and Salinity of
Fluid Inclusions
Fluid inclusion data of quartz indicate that
the Se- and Te-types formed over temperature
ranges between 160 - 330C and 160 - 350C, on
the average (shallower to deeper) of around 170 220C and 190 - 270C, respectively. The salinity
of ore fluids for the Se-type is estimated to have
been slightly lower than that for ore fluids of the
Te-type. The Se-type has the salinity up to 3.4
wt% NaClequiv. on the average of less than 1 wt%
NaClequiv. except for the Cirotan deposit which
is up to 7.15 wt% NaClequiv. (Milesi et al., 1993)
and for Te-type is in the range of 0.2 - 4.3 wt%
NaClequiv. on the average of ~2 wt% NaClequiv.. Formation temperature and salinity estimated from
the fluid inclusion homogenization temperature
and melting temperature for both types of deposits
are summarized in Figure 4.
77

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 71-81

Se-type
Pongkor:

120

Homogenization Temperature ( C)

160

200

240

280

320

Quartz
Carbonate

360

Se-type
Pongkor:

0.0

Salinity (wt% NaClequiv.)


1.0

2.0

3.0

4.0

5.0

Quartz

Cikidang:

Cikidang:

Quartz

Quartz

Cibaliung:

Cibaliung:

Quartz

Quartz

Cisungsang:

Cisungsang:

Siliceous breccia
Precious breccia
Calcite

Siliceous breccia
breccia

Cirotan:

Cirotan:

Quartz

Sphalerite
Quartz

Te-type

Te-type

Arinem:

Arinem:

Quartz
Calcite
Sphalerite

Quartz
Calcite
Sphalerite

Cineam:

Cineam:

Quartz vein

Quartz vein

Figure 4. Temperature and salinity ranges of the Se- and Te-type deposits estimated from homogenization and melting
temperatures of fluid inclusions.

Discussions

IJ
O

Mineralogically, sulfide minerals are abundant


in the Te-type and varied in the Se-type from trace
(e.g. Pongkor) to abundant (e.g. Cirotan). This
phenomenon is contradictory with the Se- and Tetypes in Japan where sulfide minerals, except for
pyrite and marcasite are very poor in amount for
the Te-type, but sulfide minerals such as argentite,
sphalerite and galena are abundant in the Se-type
(Shikazono et al., 1990).
The Se- and Te-type deposits in western Java
are characterized by the temperature generally
of <300C, on the average of less than 220C for
Se-type and less than 270C for Te-type. These
also show a general decrease of temperature with
decreasing depth and increasing paragenetic time.
The salinity is low (frequently <4 wt% NaClequiv.)
and the fluids probably are meteoric in origin with
some sources being mixed with the magmatic
origin.
The characteristics of Se-type deposits worldwide shows that electrum and selenide minerals
have been deposited from fluid with temperature
ranging around 150 - 210C (Matsuhisa et al.,
1985; Izawa et al., 1990; So et al., 1995) as indicated by fluid inclusions with low salinity (0.4
- 1.6 wt% NaClequiv.). The fluids are meteoric in
origin (Matsuhisa et al., 1985; So et al., 1995) al-

though magmatic solutions may have mixed with


meteoric water such as in Hishikari (Matsuhisa
and Aoki, 1994).
Otherwise, the temperatures of telluride
deposition were less than 354C (melting point
of hypogene sylvanite) in general and usually are
below 250C with salinities in the range of <1
to 6 wt% NaClequiv. (Kelly and Goddard, 1969;
Saunders and May, 1986; Ahmad et al., 1987).
Textural evidence shows that tellurium mineral
was formed after sulfide mineral and this concludes that magmatic sulfur was put to the fluid
first then followed by input of tellurium. Some
sulfide minerals also sometimes contain tellurium
concentration due to minor substitution of Te for
S. Sindeeva (1964) cites tellurium will presumably be concentrated in sulfide or oxide melts or
in an aqueous phase evolved from crystallizing
magmas due to the apparent rejection of tellurium
by silicate minerals.
Shikazono et al. (1990) reported that Te-type
deposits in Japan occurred at a higher level than
Se mineralization in the same mining district.
Aoki (1988) investigated the Osorezan hot springs
and at the very shallow levels he found Au-Te
minerals (e.g. krennerite, coloradoite) but no Se
minerals have been identified. In the western Java
mineralization, some Te mineralizations are associated with the Se-type deposits (Pongkor and

78

Epithermal Gold-Silver Deposits in Western Java, Indonesia: Gold-Silver Selenide-Telluride Mineralization


(E.T. Yuningsih et al.)

the Te-mineralization probably occurred closer


to the volcanic centre and at a higher level of the
geothermal system than the Se-mineralization.
It is also concluded that there might be other
factors controlled the formation of the Se- and
Te-minerals within those deposits.

Acknowledgement
The authors would like to thank PT. Antam Tbk.
for support access to data and samples during the
field investigation and to acknowledge the contribution of the large member of geologic staff.
This work is funded by the Directorate General
for Higher Education (DGHE), Ministry of Education, Indonesia, and the Faculty for the Future
Program (FFTF) Schlumberger, France.

Cibaliung), though Se minerals are uncommon


to coexist with the Te-type deposit. Comparison
with other Se- and Te-types in Japan pointed the
physicochemical conditions of the Arinem and
Cineam deposits exhibited mineral assemblages
might be deposited closer to the heat source and
shallower than those of the western most deposits
(Pongkor, Cikidang, Cibaliung, Cisungsang, and
Cirotan).
Geochemical analysis of Se and Te elements
from both Se-type of Pongkor and Te-type of
Arinem deposits show the content of Te is higher
at the vein samples of Te-type Arinem deposit,
but the difference of Se content from both types
of deposits is not too significant (Table 2). Otherwise, the petrographic investigation shows the
occurrence of Se-mineral at Pongkor, but not
at Te-types of Arinem and it is rare in Cineam
deposits. Thus, it is concluded that there are
other factors besides the host rock types, and
the distance from the heat source controlled the
formation of the Se- and Te-minerals among the
Se- and Te-type deposits.

IJ
O

References

Conclusions

The ore mineralogy of the Te-type deposits of


western Java was characterized by the abundance
of sulfide minerals with minor Te-minerals of
hessite, petzite, stutzite, tetradymite, and altaite,
while the Se-type has various amounts of sulfide
minerals with the occurrence of minor Se-minerals of aguilarite and naumannite, and Se-bearing
minerals of argentite, polybasite, and pyrargyrite.
Other minerals were found as minor or trace in
both types of deposits.
The mineralogic data indicate that the Se- and
Te-type deposits in western Java are characterized
by the presence of a large amount of quartz and
carbonates, with accessories of illite, chlorite, and
smectite. Adularia is present at the Se-type but not
in the Te-type, and generally the propylitic and
argillic alteration zonation of the Se- and Te-types
is similar. Formation temperatures of the Te-type
are generally higher than those for the Se-type.
The comparison with the Se- and Te-types
occurred in Japan pointed to the conclusion that

Ahmad, M., Solomon, M., and Walshe, J.L., 1987.


Mineralogical and geochemical studies of the
Emperor gold telluride deposit, Fiji. Economic
Geology, 82, p.345-370.
Angeles, C.A., Prihatmoko, S., and Walker, J.S.,
2002. Geology and alteration-mineralization
characteristics of the Cibaliung epithermal
gold deposit, Banten, Indonesia. Resource
Geology, 52, p.329-339.
Aoki, M., 1988. Gold mineralization in the Osorezan hydrothermal system. Rock alteration
and hot spring precipitates. Mining Geology,
38, p.64.
Basuki, A., Sumanagara, D.A., and Sinambela,
D., 1994. The Gunung Pongkor gold-silver deposit, West Java, Indonesia. In: van Leeuwen,
T.M., Hedenquist, J.W., James, L.P., and Dow,
J.A.S. (eds.), Indonesian mineral deposits:
Discoveries of the past 25 years. Journal of
Geochemical Exploration, 50, Amsterdam,
p.371-391.
Bodnar, R. J., 1993. Revised equation and table
for determining the freezing-point depression
of H2O-NaCl solution. Geochimica et Cosmochimica Acta, 57, p.683-684.
79

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 71-81

tion of the Hishikari epithermal gold-silver


veins, southern Kyushu, Japan. Economic
Geology, 89, p.1608-1613.
Matsuhisa, Y., Morishita, Y., and Sato, T., 1985.
Oxygen and carbon isotope variations in goldbearing hydrothermal veins in the Kushikino
mining area, southern Kyushu, Japan. Economic Geology, 80, p.283-293.
Milesi, J.P., Marcoux, E., Nehlig, P., Sunarya,
Y., Sukandar, A., and Felenc, J., 1993. Cirotan, West Java, Indonesia: A 1.7 Ma Hybrid
Epithermal Au-Ag-Sn-W Deposit. Economic
Geology, 89, p.227-245.
Milesi, J.P., Marcoux, E., Sitorus, T., Simandjuntak, M., Leroy, J., and Baily, L., 1999.
Pongkor (West Java, Indonesia): A Pliocene
supergene-enriched epithermal Au-Ag- (Mn)
deposit. Mineralium Deposita, 34, p.131-149.
Rosana, M.F., Haryanto, A.D., Yuniardi, Y.,
and Yuningsih, E.T., 2006. The occurrences
of base metal Mineralization in CikaduCisungsang area, Banten Province, Indonesia.
Prosiding Persidangan Bersama Geosains
ITB-UKM, p.164-166.
Rosana, M.F. and Matsueda, H., 2002. Cikidang
hydrothermal gold deposit in western Java,
Indonesia. Resource Geology, 52, p. 341-358.
Saunders, J.A. and May, E.R., 1986. Bessie G: A
high-grade epithermal gold telluride deposit,
La Plata County, Colorado, U.S.A. In: Macdonald, A.J. (ed.), Proceedings of Gold 86,
An International Symposium on the Geology
of Gold: Toronto, p.436-444.
Shikazono, N., Nakata, M., and Shimizu, M.,
1990. Geochemical, mineralogic and geologic
characteristics of Se- and Te-bearing epithermal gold deposits in Japan. Mining Geology,
40, p.337-352.
Sillitoe, R.H. and Hedenquist, J.W., 2003. Linkages between volcano-tectonic setting, orefluid compositions, and epithermal precious
metal deposits. Economic Geology, Special
Publication, 10, p.315-343.
Sindeeva, N.D., 1964. Mineralogy and types of
deposits of selenium and tellurium. Interscience publishers, 363pp.
So, C.S., Dunchenko, V.Y., Yun, S.T., Park, M.E.,
Choi, S.G., and Shelton, K.L., 1995. Te- and

IJ
O

Carlile, J.C. and Mitchell, A.H.G., 1994. Magmatic arcs and associated gold and copper
mineralization in Indonesia. In: van Leeuwen,
T.M., Hedenquist, J.W., James, L.P., and Dow,
J.A.S. (eds.), Indonesian mineral deposits:
Discoveries of the past 25 years. Journal of
Geochemical Exploration, 50, Amsterdam,
p.91-142.
Harijoko, A., Ohbuchi, Y., Motomura, Y., Imai,
A., and Watanabe, K., 2007. Characteristic
of the Cibaliung gold deposit: Miocene lowsulfidation-type epithermal gold deposit in
western Java, Indonesia. Resource Geology,
57, p.114-123.
Harijoko, A., Sanematsu, K., Duncan, R.A., Prihatmoko, S., and Watanabe, K., 2004. Timing
of the mineralization and volcanism at Cibaliung gold deposit, western Java, Indonesia.
Resource Geology, 54, p.187-195.
Izawa, E., Urashima, Y., Ibaraki, K., Suzuki, R.,
Yokoyama, T., Kawasaki, K., Koga, A., and
Taguchi, S., 1990. The Hishikari gold deposit:
high grade gold epithermal epithermal veins
in Quaternary volcanics of southern Kyushu,
Japan. Journal of Geochemical Exploration,
36, p.1-36.
Kelly, W.C. and Goddard, E.N., 1969. Telluride
ores of Boulder County, Colorado. Memoir of
Geological Society of America, 109, 237pp.
Marcoux, E. and Milesi, J.P., 1994. Epithermal
gold deposit in West Java, Indonesia: Geology, age and crustal source. In: van Leeuwen,
T.M., Hedenquist, J.W., James, L.P., and Dow,
J.A.S. (eds.), Indonesian mineral deposits:
Discoveries of the past 25 years. Journal of
Geochemical Exploration. 50, Amsterdam,
p.393-408.
Marcoux, E., Milesi, J.P., Soeharto, S., and
Rinawan, R., 1993. Noteworthy mineralogy of
the Au-Ag-Sn-W(Bi) epithermal ore deposit
of Cirotan, West Java, Indonesia. Canadian
Mineralogist, 31, p.727-744.
Marjoribanks, R., 2000. Geology of the HonjeCibaliung Area, Indonesia - An air photo
interpretation based study. Unpublished Report, 13pp.
Matsuhisa, Y. and Aoki, M., 1994. Temperature
and oxygen isotope variations during forma-

80

Epithermal Gold-Silver Deposits in Western Java, Indonesia: Gold-Silver Selenide-Telluride Mineralization


(E.T. Yuningsih et al.)

eralogic, fluid inclusion microthermometric


and stable isotope data of the Ciurug-Cikoret
veins. Resource Geology, 57, p.136-148.
Warmada, I.W., Lehmann, B., and Simandjuntak,
M., 2003. Polymetallic sulfides and sulfosalts
of the Pongkor epithermal gold-silver deposit,
West Java, Indonesia. Canadian Mineralogist,
41, p.185-200.
Warmada, I.W., Lehmann, B., Simandjuntak, M.,
and Hemes, H.S., 2007. Fluid inclusion, REE
and stable isotope study of carbonate minerals from the Pongkor epithermal gold-silver
deposit, West Java, Indonesia. Resource Geology, 57, p.124-135.
Widi, B.N. and Matsueda, H., 1998. Epithermal
gold-silver-tellurides- deposit of Cineam,
Tasikmalaya District, West Java, Indonesia.
Directorate of Mineral Resources Indonesia,
Special Publication, 96, p.1-19.
Yuningsih, E.T., Matsueda, H., Setyaraharja,
E.P., and Rosana, M.F., 2012. The Arinem
Te-bearing gold-silver-base metal deposit,
West Java, Indonesia. Resource Geology, 62,
p.140-158.

IJ
O

Se-bearing epithermal Au-Ag mineralization,


Prasolovskoye, Kunashir Island, Kuril Island
arc. Economic Geology, 90, p.105-117.
Sudana, D. and Santosa, S., 1992. Geology of
the Cikarang Quadrangle, Java. Scale 1:
100.000. Geological Research and Development Centre, 13pp.
Sukarna, D., 1999. Rare elements distribution in
Cirotan epithermal gold deposits. Indonesian
Mining, 5, p.1-10.
Sukarna, D., Noya, Y., and Mangga, S.A., 1994.
Petrology and geochemistry of the Tertiary
plutonic and volcanic rocks in the Bayah area.
Proceedings of Indonesian Geologist Association, p.389-412.
Syafrizal, Imai, A., Motomura, Y., and Watanabe,
K., 2005. Characteristics of gold mineralization at the Ciurug vein, Pongkor gold-silver
deposit, West Java, Indonesia. Resource Geology, 55, p.225-238.
Syafrizal, Imai, A., and Watanabe, K., 2007. Origin of ore-forming fluids responsible for gold
mineralization of the Pongkor Au-Ag deposit,
West Java, Indonesia: Evidence from min-

81

Indonesian Journal on Geoscience Vol. 1 No. 2 August 2014: 83-97

INDONESIAN JOURNAL ON GEOSCIENCE


Geological Agency
Ministry of Energy and Mineral Resources
Journal homepage: hp://ijog.bgl.esdm.go.id
ISSN 2355-9314 (Print), e-ISSN 2355-9306 (Online)

Reservoir Modeling of Carbonate on Fika Field:


The Challenge to Capture the Complexity of Rock and Oil Types
Erawati Fitriyani Adji1, Febrian Asrul2, M. Aidil Arham3, and Bayu Wisnubroto4
Petrophysicist, PT Medco E&P Indonesia
Reservoir Engineer, PT Medco E&P Indonesia
3
Development Geologist, PT Medco E&P Indonesia
4
Geologist Software Support, Schlumberger
1

Corresponding author: era_206@yahoo.com


Manuscript received: October 10, 2013, revised: January 21, 2014, approved: August 12, 2014

IJ
O

Abstract - The carbonate on Fika Field has a special character, because it grew above a basement high with the thickness and internal character variation. To develop the field, a proper geological model which can be used in reservoir
simulation was needed. This model has to represent the complexity of the rock type and the variety of oil types
among the clusters. Creating this model was challenging due to the heterogeneity of the Baturaja Formation (BRF):
Early Miocene reef, carbonate platform, and breccia conglomerate grew up above the basement with a variety of
thickness and quality distributions. The reservoir thickness varies between 23 - 600 ft and 3D seismic frequency
ranges from 1 - 80 Hz with 25 Hz dominant frequency. Structurally, the Fika Field has a high basement slope, which
has an impact on the flow unit layering slope. Based on production data, each area shows different characteristics and
performance: some areas have high water cut and low cumulative production. Oil properties from several clusters
also vary in wax content. The wax content can potentially build up a deposit inside tubing and flow-line, resulted
in a possible disturbance to the operation. Five well cores were analyzed, including thin section and XRD. Seven
check-shot data and 3D seismic Pre-Stack Time Migration (PSTM) were available with limited seismic resolution.
A seismic analysis was done after well seismic tie was completed. This analysis included paleogeography, depth
structure map, and distribution of reservoir and basement. Core and log data generated facies carbonate distribution
and rock typing, defining properties for log analysis and permeability prediction for each zone. An Sw prediction for
each well was created by J-function analysis. This elaborates capillary pressure from core data, so it is very similar to
the real conditions. Different stages of the initial model were done i.e. scale-up properties, data analysis, variogram
modeling, and then the properties were distributed using the geostatistic method. Finally, after G&G collaborated
with petrophysicists and reservoir engineers to complete their integrated analysis, a geological model was finally
created. After that, material balance was needed to confirm reserve calculations. The result of OOIP (Original Oil in
Place) and OGIP (Original Gas in Place) were confirmed, because it was similar to the production data and reservoir
pressure. The model was then ready to be used in reservoir simulation.
Keywords: reservoir modeling, carbonate, rock and oil types, simulation, Fika Field

Introduction

Fika Field is an oil and gas producer that lies


in South Sumatra Basin. Currently, the field has
38 wells, of which 24 are producers from BRF
(Baturaja Formation). The formation has heterogenic properties. Some parts of the field have
BRF with high permeability, while the other
parts may have BRF with tight permeability that
requires stimulation, such as hydraulic fracture,

in order to be able to produce. The cumulative


production is 8 MMSTB and more than 47
BCF of gas, which originated from associated
gas and gas cap production. Oil recovery factor
is expected to be more than 30%, although the
gas cap has been blow- down since December
2009, which has accelerated reservoir pressure
depletion and reduced oil production. In addition, hydraulic fracturing has been done in this
field, resulting in an increase in oil production

IJOG/JGI (Jurnal Geologi Indonesia) - Acredited by LIPI No. 547/AU2/P2MI-LIPI/06/2013, valid 21 June 2013 - 21 June 2016

83

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 83-97

from exinite, liptinit or algae. This generally


indicates gas and oil. Lemat Formation on the
studied area began 22 MYA and has moderate
maturity for producing oil (early oil generation)
in Benakat Gully.
Data and Method
This research was divided into various
stages of data analysis, as listed below:

Seismic Data Analysis


1. Well seismic tie from seventhcheck-shots.
2. Seismic interpretation and the result as time
structure and depth structure maps of reservoir and basement.

from 20 BOPD to 50 - 113 BOPD, while, other


wells produce gas and water.
Because a high demand for gas must be satisfied, the Fika Field must produce its gas cap,
which is the main reservoir drive. This will affect
reservoir pressure and oil recovery. To minimize
oil loss due to gas cap blow down, and to maximize gas production, a team was established to
conduct a reservoir study.
The previous workers who studied Baturaja
carbonates relating to hydrocarbon reservoir
properties are Caroline (2005), Handayani (2008),
and Erawati (2013).
The purpose of this paper is to explain how
to build rock typing from carbonate which is
highly heterogenic, and how to generate permeability transform and steps in model water
saturation by using capillary pressure from core
analysis. At the end of the paper, there is a discussion on reserve confirmation regarding static data
and production data by utilizing material balance.

Petrophysical Evaluation
1. Review available Special Core Analysis
(SCAL) data to determine of a, m, n.
Facies carbonate assignment after core
depth matching and core description.
Net Overburden (NOB) core correction for
porosity and permeability and Klinkenberg correction for permeability.
Defining matrix end-point value from
crossplot: RHOB vs core porosity, DT vs.
core porosity, NPHI vs. core porosity.
Defining a and m from the best straight line
plot log F (Formation Resistivity Factor)
vs. log porosity on every facies (rocktyping result).
Defining n from the slope of the line plot
log Sw vs. log RI (Ro/Rt).
2. Analyzing log using zonation based on geological correlation.
Estimating Rw value for Baturaja reservoir in Fika field.
Calculating Vcl (mudstone) using SP log
and Density- Neutron log after confirmation with XRD data correlation.
Calculating porosity effective and Sw using Simanduox and Indonesia. The final
selection for Sw values will be based on
transition zone analysis (TZA).
Crossplot between log porosity (effective
and total) vs. core porosity (NOB correction).
3. Lithofacies based on core description and
rocktyping determination.

84

This field has a simple geological structure and


there is more emphasis on stratigraphic aspects.
Musi Platform is bounded by Pigi depression in
the northern area, Lematang depression in the
south-east area, Saung Naga graben in the southwest area, and Benakat Gully in the eastern
area. This setting indicates the possibility of reef
build-up above basement high (Musi Platform),
when the sea level rose (transgression) during
deposition of Baturaja Formation (Rashid et al.,
1998). The carbonate type that grows on the Musi
Platform is an isolated platform. Carbonate facies
on the Fika Field is divided into reef, platform,
and breccia conglomerates with different quality,
uneven distribution, and relatively thin thickness
(up to 20 ft below). The Baturaja carbonate is
Early-Middle Miocene in age with depositional
environment about neritic to shallow marine,
while tectonic settings are in a sagging phase. In
the study conducted with LAPI ITB (2011), the
Musi Platform has hydrocarbon source rock from
Lemat Formation as lacustrine environment. The
lithology is lacustrine shale mixing between algal
lacustrine and organic material from land origin.
Oil expelled on moderate maturity (approximately
0.7 - 0.95% Ro) with kerogen type II/III derived

IJ
O

Geological Setting

Reservoir Modeling of Carbonate on Fika Field:


The Challenge to Capture the Complexity of Rock and Oil Type (E.F. Adji et al.)

4. Analyzing looping log using parameter zonation by rocktype.


5. Prediction permeability and Analyzing Transition Zone (TZA) to predict Sw based on core
data.

Fine Grid Model


1. Scaling up properties
2. Analyzing data
3. Modeling variogram
4. Distributing the properties using geostatistic
method.

Reserve Calculation Confirmation


1. Calculation material balance reserve
2. Calculation static model reserve (OOIP and
OGIP).

Match value between synthetic seismogram and


seismic trace is called coeficient correlation (r).
Positive r value and near with 1 shows that synthetic seismogram and seismic trace has good
correlation.
Well to seismic tie analysis was done on the
seventh check shot at this field. Figure 2 shows
synthetic seismogram from Fika-1 well, while
Table 1 shows the resume of coeficient correlation
from each well.
After well seismic tie, some main markers
were defined and distributed on seismic data to
obtain the geological model of each marker and
to interpret the geological history of the field.
The seismic mapping result of basement and
Baturaja carbonate is shown in Figures 3a and
3b, respectively.

Result and Discussion

IJ
O

Seismic Data Analysis


Seismic 3D PSTM was used for this study.
Well seismic tie was implemented in the early
stages of seismic well analysis, using well data
(density and sonic log) and wave model (wavelet)
from seismic data extraction. The same parameters
as 3D seismic data were used, where positive polarity is recorded as increasing acoustic impedance
on positive amplitude with zero phase. Figure 1
below shows wavelet extraction parameter, the
model of extraction, and the amplitude spectrum.
The next step was to create a synthetic seismogram and to match the trace with seismic data.

Figure 1. Parameter and the result extraction on Field 3D


seismics.

Figure 2. Parameter and the result extraction on Field 3D


seismics.

Table 1. Resume of Coeficient Correlation from each Fikas


check Shot Wells
Well

Coefficient Correlation (r)

Fika-1

0.605

Fika-2

0.698

Fika-3

0.781

Fika-4

0.281

Fika-5

0.447

Fika-6

0.463

Fika-7

0.845

Petrophysical Evaluation
This step begins with an analysis of core
data measurement after core depth matching. It
is important to make a reliable definition of the
position of the carbonate facies development
with depositional setting and match with the
subsurface condition. Thereafter, routine core
85

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 83-97

perm NOB vs amb

10000

Fika-07

Perm at NOB

1000

Fika-06

Fika-01

Fika-02

Fika-03

300

600

1.0288

10
1
0.1

0.01

Fika-05

Fika-04

y = 0.5159x

100

900

1200

1500 m

0.001
0.01

0.1

10

100

Perm at Ambient

1000

10000

Fika-07

Figure 5. Permeability NOB correction correlation on Fika


Filed.

Fika-06

Fika-01

Fika-02

300

600

900

1200

0.8

1500 m

Figure 3. Time Structure Map. (a) Basement Fika, (b)


Baturaja Carbonates.

0.6

y = -0.0024x + 0.013x + 0.0606x + 0.6267


3

0.4

Trendline

-1

por NOB vs amb

Porosity at NOB

40
35

y = 0.9801x

30
25
20
15
10
5
0

10

20

Porosity at Ambient

30

Figure 4. Porosity NOB correction on Fika Field.

86

40

Log kair

BRF at Fika field was carried out in a previous


study which indicated that seven lithofacies can
be identified based on core calibration from FikaA1, C1, D1, E1, and F2 wells, in addition to image
analysis from Fika-B4, C1, and E1 wells.
The G&G groups utilized the available seismic data and well logs to define three depositional
facies (referred to as zones in the current geologic
model) as follows:
1. Reef Limestone
2. Platform Limestone
3. Breccia/Conglomerate Clastics
Investigation of available core description (both
whole & plugs) confirms the existence of the
above three depositional facies and indicates the
following lithofacies (Figures 7 and 8; and Table 2).
The distribution of lithofacies described
above (from core data) does not indicate any
specific relationship with subsea elevation (TVD
subsea) within individual depositional facies
(zones) as shown in Figure 9.
The above conclusion is supported by the
previous study as indicated by the distribution
of lithofacies with depths shown in Figure 10.

Klinkenberg effect on permeability is estimated from available liquid permeability data and
given trend from text book (Figure 6).
Based on geological setting, the carbonate
facies which developed from the top of Baturaja
to basement are reef, platform, and breccia/
conglomerate clastics. A previous study on the

Figure 6. Klinkenberg correction correlation of permeability


on Fika Field.

1.0288

kNOB = 0.5159 kamb

Text Book Data


Actual Soka Data

NOB = 0.9755 amb

0.2

IJ
O

analysis (RCA) and SCAL data were done.


Porosity core requires NOB correction (Figure 4),
while permeability core requires NOB (Figure 5)
and Klinkenberg correction (Figure 6). Based on
the core description, facies carbonate definition is
created and zonation is needed for log analysis.
The correction factors are as below:

Correction Factor

Klinkenberg Effect

Fika-05

Fika-04

Fika-03

Reservoir Modeling of Carbonate on Fika Field:


The Challenge to Capture the Complexity of Rock and Oil Type (E.F. Adji et al.)

Core Calibration Lithofacies


2

Conglomerate/breccia

Mudstone

Limestone

Massive to
Volcanic
Mottled
microcrystalline Mudstone conglomerate
wackstone
wackstone

Volcanic
breccia

BeddedSkeletal packlaminated stone to grainwack to pack


stone

Figure 7. Core calibration lithofacies.

LF-2

LF-3

LF-4

IJ
O

LF-1

Vuggy to mottled LS

Bedded to lamited LS

Massive to
microcrystalline LS

Inregular layers to mottled LS

LF-6

LF-5

LF-7

Equal bedded to wavy LS

Mudstone

Rubble bed
(Breccia/conglomerate)

No core for calibration


(Well-D, 3337-3341)

Figure 8. Image and core calibration lithofacies.

Table 2. Facies Distribution from Core Analysis on Fika


Field
Lithofacies

Number of Percentage
Data Points

Reef

Vuggy Coral/Grainstone
Mottled Wackstone

34
36

11%
12%

Platform

Vuggy Wackstone/Packstone

31

10%

Bedded, Chalky, and microcrystalline


Limstone

109

36%

Diminant Limestone Fragments

60

20%

Dominant Vulcanic/Basement Fragments


Total

31

10%

301

100%

Breccia/Conglomerate

2850

Depth TVDss

Depositional
Facies

2800

2900

Rock Type

Reef Vuggy
Reef non Vuggy
Platform Vuggy
Platform Non Vuggy
Breccia Limestone
Breccia Clastic

2950
3000
3050
3100
3150

Figure 9. Relationship between lithofacies and subsea elevation on Fika Field.

87

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 83-97

Legend:

Well-B

LF-1 : Bedded to laminated LS


LF-2 : Vuggy to mottled LS

MDI-TVD

Grain size

LF-3 : Inregular layers to mottled LS

3215/-2885.5

Well-D
Grain size

Well-6-2
Grain size

MDI-TVD

MDI-TVD

3285/-2946.63

LF-4 : Massive to micro-crystalline LS


LF-5 :Mudstone
LF-6 :Rubble bed
LF-7 : Equal to wavy bedded LS

Well-C

MDI-TVD

Grain size

3266/-2958

10 ft

Well-E

MDI-TVD

Grain size

3244/-2936.55

3355/-3029.5 Basement

3467/-3133.63 Basement

3297/-2996 Basement

3238/-2980.55 Basement

3320/-3011 Basement

Figure 10. Relationship between lithofacies and subsea elevation on Fika Field.

Table 3. Lithofacies Code


Lithofacies
code

Description

Nonreservoir (no log above basement)

Breccia/Conglomerate with volcanic

IJ
O

The facies distribution shown in the above


chart indicates that some thin intervals of breccia/
conglomerate exist within the limestone depositional facies. This phenomenon is applied in the
current geologic model, where only three zones
are included, as discussed above. If this situation
is not acceptable, some consideration should be
given to include the breccias lithofacies (with
limestone fragments and with volcanic/basement fragments) in the lithofacies distribution
of limestone zones and assigning appropriate
percentages to represent the thin breccia intervals
within reef and platform zones.
It should be noted that the mudstone lithofacies (defined as LT-5 in the previous study) is not
recognized in any core plugs, but is included in
the whole core description.
Accordingly, the present model will include
these lithofacies as a result of the cut-off analysis
using appropriate porosity, Vms, and permeability
cut-off values and will be referred to as nonreservoir facies (facies 0 in Petrel).
Based on the above discussion, the following
facies code (rock type) is defined for the geologic
model, was shown on Table 3.
The following lithofacies assignment (by
zone) will be utilized in the model if breccia/conglomerate thin interval within reef and platform
are ignored (Table 4).
After lithofacies were created on the cored
wells, it was necessary to distribute the lithofacies
on all uncored wells. Although it was difficult to

88

Breccia/Conglomerate with limestone


fragments

Bedded, Chalky, and microcrystalline


Limestone

Vuggy Wackstone/Packstone

Mottled Wackstone

Vuggy Coral/Grainstone

Table 4. Zone Code and Lithofacies


Zone
Code

Description

Lithofaces
included

Reef

0, 5 and 6

Platform

0, 3 and 4

Breccia/Conglomerate

0, 1 and 2

obtain the special log character of each lithofacies, it could be done using crossplots between
RHOB - NPHI, RHOB - DT, RHOB - PHIT,
RHOB - SP, RHOB - PHIE, and RHOB - GR.
Consequently, the lithofacies needed to be simplified in order to distribute on uncored (electrofacies) wells, as shown below (Table 5).
Low energy limestone represents mud dominated on rock matrix, while high energy carbonate
represents grain dominated on rock matrix. The

Reservoir Modeling of Carbonate on Fika Field:


The Challenge to Capture the Complexity of Rock and Oil Type (E.F. Adji et al.)

The next step was to define matrix end-point


value from crossplot between log data and core
data measurement, i.e. RHOB vs. core porosity,
DT vs. core porosity, NPHI vs core porosity (Table
7). Logically, when porosity value is zero, it is
assumed as matrix value on the log data.
The preliminary log analysis used zonation
based on geological correlation, after rock-typing
had been defined. Log analysis uses parameter a,
m, n, and end-point matrix in every rock type.
Rw estimation for Baturaja reservoir was based
on Picket Plot (Figure 13). The Rw estimation
value was equal to 17.000 ppm salinity. A water
test lab analysis result was not appropriate input
for log analysis, due to the influence of mud on
the water samples.
Vcl (mudstone) calculation on Fikas carbonate used GR and density-neutron log. According
to the concept introduced by Asquith (2004), the
sonic log usually reads matrix porosity without the
effect of vugs, but both neutron and density logs
indicate the effect of vugs on porosity reading.
Consequently, more significant differences were
expected between calculated porosity values from
these logs opposite vuggy limestone intervals
compared to intervals without vugs. The following
chart shows this phenomenon after applying the
concept to Fikas core data (Figure 14).
In the above chart, the porosity difference ratio
is defined as follows:

Table 5. New Zone Code and Lithofacies


New zone
code

Lithofaces

Description

included

Non reservoir

Breccia/Conglomerate

0, 1 and 2

Low energy LS

0, 3

High energy LS

0, 5

Vuggy LS

0, 4, 6

IJ
O

electrofacies distribution was based on the value


of VGR and corrected SP. After reviewing all
capillary pressure data, it has been concluded that
the permeability/porosity ratio depends more on
rock type from lithofacies and capillary pressure
data. The permeability/porosity ratio will be used
as a basis to create rock typing and TZA. Rock
typing based on permeability/porosity ratio and
TZA will be discussed later in this paper in a
special section.
Based on SCAL data from four cored wells,
a and m were defined from the best straight line
plot log F (Formation resistivity factor) vs log
porosity on each rock type (Figure 11). The n
parameter was defined from the slope of the line
plot log Sw vs log RI (Ro/Rt) (Figure 12). The
average density was defined on each rock type.
The result is shown in the Table 6.

0.1

10

y = 1.0768x

-1.795

Porosity, v/v

RT 1

a and m from Rock Type 2

100

Formation Factor

Formation Factor

a and m from Rock Type 1

y = 1.0995x

0.1

RT 2

Formation Factor

a and m from Rock Type 3

0.1

-1.761

Porosity, v/v

Power (RT1)

100

10

Power (RT2)

100

10

y = 0.702x

-2.16

Porosity, v/v
RT 3

Power (RT3)

Figure 11. Formation Factor vs. Porosity to obtain Cementation Factor (m).

89

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 83-97

n from Rock Type 1

n from Rock Type 2

1.5

1.5

1.4

1
0.8
0.6
0.4
0.2
-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

log Water Saturation


RT 1

-0.1

0
0

1.4
1.2

y = -1.9357x

log Resistivity Index

log Resistivity Index


-0.8

1.2

y = -1.905x

1
0.8
0.6
0.4
0.2

-0.6

-0.5

-0.4

-0.3

-0.2

log Water Saturation

Linier (RT1)

RT 1

n from Rock Type 2

-0.1

0
0

Linier (RT1)

1.5
1.4

log Resistivity Index

y = -1.9588x

1.2
1
0.8
0.6
0.4

0.2

-0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1

log Water Saturation


RT 3

0
0

Linier (RT3)

Figure 12. Formation Resistivity Index vs Brine Saturation to obtain Saturation Exponent (n).

It should be noted that this definition of porosity difference ratio does not align with the results
from this study, since theoretically speaking,
sonic porosity should be lower than densityneutron porosity for interval with vugs. However,
log analysis results indicated sonic porosity to
be higher than density-neutron porosity for most
intervals.
Even with the incorrect definition, the results
in the above chart do not indicate any correlation
for defining criteria to identify vuggy intervals. It
should be noted that Vsh values initialy calculated
for this study were based on minimum values
among several methods available in the Petrolog
software. The study team revised this concept
in view of the questionable applicability of GR
logs in carbonate reservoirs. Accordingly, only
density-neutron logs were used to define Vsh.
The core data do not include platform with
vugs. Accordingly, the study team decided not to
include this rock type in the current model. This
decision was further supported by the geologic
concept of low probability for finding vugs in
platform carbonate intervals overlain by reef
carbonate.
The fact that vuggy intervals that cannot be
recognized from well logs is supported by visual
investigation of available core material. Figure
15 shows that vugs were scattered within thin

IJ
O

Table 6. Resume of Average Value from a, m, n, and Grain


Density on every Rock Type
Rock Type
1
2
3

Rock Type
1
2
3

From Cross Plot


a
m
1.077
1.795
1.100
1.761
0.702
2.160

n
1.905
1.936
1.959

Ave Grain Dens, gr/cc


2.705
2.692
2.699

Table 7. Resume of Average Value from a, m, n, and Grain


Density on every Rock Type
Parameter
Matrix density
Fluid density

NPHI for matrix


NPHI for fluid

RT1

RT2

RT3

2.68

2.7

2.71

1.1

1.2

1.1

0.07

0.83

0.94

1.1

Matrix transit time

50

52

53

Fluid transit time

225

198

210

R = sonic - D-N
D-N

Where:
sonic = sonic porosity
D-N = average porosity from density and
neutron logs
90

Reservoir Modeling of Carbonate on Fika Field:


The Challenge to Capture the Complexity of Rock and Oil Type (E.F. Adji et al.)
RT vs PHIAvs VCL
Fika A1 (S) pro log data: Zone 5 - 5260.000 Ft to 5512.5000 FT

PHIA (v/v) - Apparent Porosity


(from PHICP, PHID, PHIS, PHIN, or PHIM)

1000

0.100

0.010
0.2

20

200

2000

RT (OHMM) - Formation Resistivity


0

VCL (v/v)

Figure 13. Picket plot to determine Rw of Baturaja Formation.

Porosity difference ratio for various rock types

1
0.5
0

IJ
O

Porosity diffence ratio

1.5

existence of an appreciable amount of mudstone


within the BRF, including the reef zone. The
mudstone in BRF is believed to be the result of
internal diagenetic and lithification effects and
probably some external effects from gravity
settling of fine materials.
Three thin sections were analyzed quantitavely by Lemigas in order to determine the
mudstone content in the side wall samples from
well Fika-I1 (1). The results are shown below.
Measured mudstone contents in the three sections are 15, 16 and 62.5% by volume. Average
Vsh value for the sorted data sample is 21.6%,
which is rather low for the mudstone content
range indicated by thin section analysis.
After defining appropriate Vsh, total porosity
must be correctly calculated to be effective porosity. Comparative results between log porosity
(total and effective) vs. core porosity (NOB) are
shown in Figure 16.
Sw calculation was made using Simanduox
and the Indonesia method. The Indonesia method is more appropriate in this field, because the
result is more sensitive to transition areas. The
final selection for Sw values was based on TZA.
Permeability transform of four lithofacies
on electrofacies was slightly modified. Vuggy
limestone has data distribution near low energy limestone, so the permeability transform
between low energy limestone (mud supported
dominated) and vuggy limestone used the same
transform value. The formula can be seen in the
Figures 17a, b, and c.

-0.5
-1

Rock Type

Figure 14. Porosity difference ratio for various rock type


on Fika Field.

Meteoric diagenesis

Dissolution of large benthic forams due to fresh


Water leaching during subaerial exposure

5mm
0
Leaching is more intense in exposed marine
limestone on topographic highs

Figure 15. Meteoric diagenesis from thin section on Fika


Field.

intervals that cannot be read and cannot affect log response. Investigation of side wall core
samples from well Fika I1(1) as well as thin
section analysis of these samples indicate the

91

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 83-97

Porosity Plot

Porosity Plot

40

40

35

35

30

30
25

Trend
A1
F2
E1
C1

20
15

PHIE

PHIE

25

15

10

10

10

15

20

25

30

Core Por NOB

35

40

Trend
A1
F2
E1
C1

20

10

15

20

25

30

35

40

Core Por NOB

Figure 16. Comparation between plot between log porosity (total and effective) vs. core porosity (NOB).

log k vs porosity for Rocktype 3


(high energy carbonate)
4

y = -40.415x2 + 29.399x - 2.3904

1
0

IJ
O

log k

Transition Zone Analysis


To define Sw in each grid, transition zone
analysis (TZA) was applied. The application was
conducted in the following procedure:
1. Defining permeability transforms and a
suitable Swc trend in terms of permeability,
which for Fika Field can be seen in the following graph (Figure 18).
2. Defining J-function derived from core data
and normalize all available data in a single
chart. The core data were divided into three
regions based on range of k/ (Table 8 and
Figure 19).
3. Defining J-max from chart in no. 2
4. Calculating k, Swc and Sw* from the exploration well log (i.e. the log which was surveyed
when the reservoir was not yet producing) and
using it to calculate h and (J cos ) for every
depth-log above OWC.
J cos = h(w- o)gk/

-1

y = 14.543x - 2.1684

-3

0.05

vuggy

0.1

grain supported

0.15

0.2

Porosity

0.25

0.35

0.4

Poly (polinomial method)

Linear (linear method)

log k vs porosity for Rt2


(low energy carbonate)

y = 16.167x - 1.9623

y = -36.591x + 27.873x - 2.4362

log k

0.3

-2

1
0

-1
-2
-3

0.05

0.1

mud supported

vuggy

0.15

0.2

Porosity

0.25

Linear (linear method)

0.3

0.35

0.4

Poly (polynomial method)

log k vs Por for Rt1


(Breccia clastics)

4.0
3.0

log k

2.0
1.0

y = 14.148x - 1.9209
R2 = 0.8757

Swc

0.0
-1.0
-2.0

0.05

0.1

0.15

0.2

Porosity

Clastics

0.25

0.3

0.35

0.4

Linear (Clastics)

Swc Transform

0.5
from Relative Permeability from Pc
0.45
0.4
0.35
0.3
0.25
0.2
0.15
4
3
2
0.1 y = -0.0007x + 0.0068x - 0.0178x - 0.052x + 0.3618
0.05
0
-1
3
1
2
0

Log (k)

Figure 17. Permeability transform for high energy limestone (a), low energy limestone and vuggy limestone (b),
and breccia clastics (c).

92

Figure 18. Crossplot between Swc vs. log permeability to


get Swc transform.

Reservoir Modeling of Carbonate on Fika Field:


The Challenge to Capture the Complexity of Rock and Oil Type (E.F. Adji et al.)
Table 8. Range Definition of Rock Typing on TZA
Lab J-function for Range 3

k/por<100
100 <k/por<1200
k/por> 1200

1.6
1.4

J eld

1.2

Log (J+1)

low k/por
medium k/por
high k/por

J lab

1
0.8
0.6
0.4
0.2
0

Lab J-function

0.1

0.2

0.2

0.4

0.3

1.4

low k/por
med k/por

1.2
1

0.6

0.7

0.8

J eld

0.4

0.8

J lab

0.2
0

0.2

0.4

0.9

0.9

0.9

1.2

0.6

0.8

Lab J-function for Range 2

high k/por

0.6

0.8

Log (J+1)

Log (J+1)

0.5

SW*

1.6

SW*

0.6
0.4
0.2

Figure 19. Crossplot between Sw* vs log (J+1) to get J


function.

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

SW*

Lab J-function for Range 1

Log (J+1)

0.6

J eld

0.5

J lab

0.4

0.3

0.2

0.1

IJ
O

5. Ploting (J cos ) versus Sw* and fitting the


best J-function curve. Calculating ( cos )
for each reservoir rock type at determined
res
value of Sw*
( cos )res = (J cos )
J
6. Calculating the reservoir J-function for each
reservoir facies using the average value of
( cos )res.
7. Comparing the curves of laboratory and
reservoir J-functions versus Sw*.The chart
should show a good match, as shown in the
following graph (Figures 20a, b, c):
8. Estimating the value of for each reservoir
rock facies if is known.
9. Calculating the coefficient of J-function for
use in Petrel model.
10. Using the reservoir J-function to formulate
a transform relating Sw* to Jres or a selected
function of Jres.
11. Using the value ( cos )res to calculate required capillary pressure curves for various
facies from their normalized J-Functions.
12. Calculating corresponding values of Sw*.

0.7

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

SW*

Figure 20. Curve shape for Lab J-function for range 3 (a),
range 2 (b), and range 1(c).

Fine Grid Model


The 3D geology model was built by using
Petrel 2011 software. There were several stages
during the process, including: 1) creating horizon, zone, and layering, 2) scaling up properties,
3) analyzing data, 4) modeling variogram, and

5) distributing the properties by geostatistic


method. The area of interest was limited by
polygon boundary, which was created based on
oil-water contact area.
Since there was no fault in this Fika Field,
the 3D grid model was built using simple grid.
The first step in creating a simple grid was creating horizon, using several surfaces as input
data, such as BRF (top surface), PLTFRM, BXCGL, and BSMT (bottom surface). All of these
input surfaces were already in depth domain.
The next step was to determine the grid boundary, which was based on surface boundary and
grid geometry, such as X minimum, X maximum, Y minimum, Y maximum, and grid size
increment. After this, the grid increment was
determined. The grid increment should represent
the geological features on a lateral distribution
and also for simulation purposes. 50x50 grid
increment was chosen, because it would give a
good result for distribution of reservoirs. If the
93

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 83-97

A histogram was used to do a comparative QC


of the well logs before and after upscale logs.
A similar trend in the histogram represented
a good correlation between well logs before
upscale and upscale logs. Data analysis was
required, because the petrophysical modeling
used a variogram from data analysis. The reason for this was to define the major, minor, and
vertical direction of several properties.
Porosity modeling was built using SGS with
conditioning to facies and subfacies for each
zone. Even though the distribution of porosity
followed the variogram, it also referred to the
AI model with applied collocated co-kriging.
So for, any area that was out of variogram, the
porosity would follow the trend of AI model
(Figure 21).
Similarly, permeability modeling was built
using SGS and conditioning to facies and subfacies, but referred to the porosity model (Figure
22).
Meanwhile, facies modeling was built using
SIS (Gslib) method, because this method was
able to distribute the discrete data very well.
The SGS (Gslib) honoured the well data and
distributed to four different facies, i.e. nonreservoir, breccias and vuggy LS, low energy LS,
and high energy LS, (Figure 23).
The other properties, vertical permeability,
net to gross, and rock type, were distributed by
applying a formula using a PETREL calculator.
The water saturation was also distributed by
using a formula obtained from the relationship
between J-function and capillary pressure based
on TZA above.

IJ
O

grid increment exceeds 50x50, there will be a


greater possibility of error when calculating
bulk volume (reflected in negative value in bulk
volume), and also the value of properties will
not be accommodated. But if less than 50x50 is
input, many active grids will be created, which
will be ineffective, and furthermore the calculation process will be slowed down. As a result,
there are four horizons with three zones. Zone
one is reef zone, zone two is platform zone, and
zone three is breccias/conglomerate zone. After
creating zonation, the following step was the
layering process. Each zone was divided into
several layers based on cell thickness, so every
layer will have an average thickness of 0.88 ft.
Determining the number of layers was based
on the depositional pattern of carbonate in Fika
Field. As a result, the layering process created
972 layers for all zones, with a total number of
5458752 3D cells.
Scale-up properties were calculated for
several well logs, i.e. porosity, horizontal permeability, vertical permeability, water saturation, facies and net to gross ratio. All of these
properties were upscaled based on the fine
grid that was created before, using weighted
averages and the following criteria: a) facies
and rock type: most of average weighted bulk
volume, b) net to gross: arithmetic average
weighted bulk volume, c) porosity: arithmetic
average weighted bulk volume, d) horizontal
permeability: arithmetic average weighted net
volume, e) vertical permeability: harmonic
average weighted bulk volume, f) water saturation: arithmetic average weighted pore volume.
Porosity

3200

3250

Soka-G1

3200

0.28
0.24
0.2
0.16
0.12
0.08
0.04

3000

00

Soka-F4

27

Soka-G2

3000

Soka-F4
Soka-F5

Soka-F2

21

50

250

Soka-D10

Soka-I2
25

00

Soka-D7

Soka-D4

Soka-A3
Soka-A1 (3)

Soka-B

Soka-B1

Soka-B/1(2)

Soka-H3
Soka-B6
Soka-B3

Figure 21. Porosity map after co-krigging stage.

94

Soka-D2ST

Soka-H4

Soka-H1
Soka-D1 Soka-D5
Soka-D

Soka-E2
0

275

Soka-D3

Soka-D9

Soka-C4
Soka-C2
Soka-H1

Soka-E1(2)
50

Soka-B2
Soka-B5

Soka-H1

32

Soka-H2

Soka-A2

Soka-F1

Soka-D6

250

500

750 m

Reservoir Modeling of Carbonate on Fika Field:


The Challenge to Capture the Complexity of Rock and Oil Type (E.F. Adji et al.)
Permeability (mD)

3200

3250

Soka-G1

3200

180
32
56
1
0.10
0.032
0.0056
0.001
0.00010

3000

Soka-F4

27

Soka-G2

3000

Soka-F4
Soka-F5

00

Soka-F2

21

50

250

Soka-D10

Soka-I2
25
00

Soka-D7

Soka-D4

Soka-A3

Soka-B5
Soka-B Soka-B/1(2)

Soka-B1

Soka-D2ST

Soka-H4

Soka-H3
Soka-B6

Soka-D

Soka-E2
0

275

Soka-D3

Soka-D9 Soka-E1(2)

Soka-C4
Soka-C2

Soka-B3

Soka-H1

50

Soka-B2

Soka-H1
Soka-D1 Soka-D5

32

Soka-A2

Soka-A1 (3)

Soka-H1

Soka-H2

Soka-F1

Soka-D6

250

500

750 m

Fika-I1(1)

Fika-A1 (3)

-2400

Subur-1

Fika-F-4

Bg = current gas FVF, RB/SCF


Bo = current oil FVF, RB/STB
Rsi = initial solution GOR, SCF/STB
Rpc = cumulative producing GOR, SCF/STB
Bti = initial two-phase FVF, RB/SCF
Bgi = initial gas FVF, RB/SCF
m = gas cap size
P = pressure, psi
C = water influx constant, RB/psi
tD = dimensionless time
QD = dimensionless flow
j = time step index for water influx constant
n = number of time steps used in water influx
calculations
From the straight line original oil in-place
in Baturaja Formation = 26 MMSTB
Water influx constant = 2500 bbls/psi
The m value of 7.3 originated from static
model (without assumption of any basement
bald). This calculation used aquifer characteristics as shown in Table 9.
The material balance calculation confirmed
the volumetric calculation that the OOIP was
around 22 - 26 MMSTB and that OGIP was
around 131 BCF.

IJ
O

-3200

-3200

-2600

Fika-D1

-2600

-2400

Figure 22. Permeability co-krigging stage.

-3600

-3600

FAC
Non Reservoar
Breccia
Low energy Lst
High Energy Lst
Vuggy Lst

Figure 23. Vertical cross section that shosw facies modeling distribution.

Reserve Calculation Confirmation


Reserve confirmation utilizing material balance
analysis
Material balance was utilized to confirm
oil and gas in-place in Fika Field. This field
has been producing since January 2001, so
there are enough production and pressure data
(Figure 24).
A straight line material balance analysis can
be seen in the graph (Figure 25).
Np [Bt + Bg (Rpc - Rsi)] + Wp
Y1

= N+C

n1 (P j-1 - Pj) QD f or tD -tDj-1


N

Y1

mBti (Bg - Bgi)


Y1 = (Bt - Bti) +
where
Bgi
and:
N = OOIP, STB
Np = cumulative oil produce, STB
Wp = cumulative water produced, bbl
Bt = current two-phase FVF, RB/SCF

Static model reserve calculation (OOIP and


OGIP)
Volumetric calculation was done to check
the original oil and gas in-place in the Fika BRF
reservoir. In addition, the original oil in place
results were compared with the material balance analysis to ensure there were similarities
between these two methods.
95

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 83-97

Conclusions

Observed Reservoir Pressure in Fika Field


From Static Bottom Hole Survey

1600
1550

Pressure, psig

1500
1450
1400
1350
1300
1250
1200
Jan-00

Jun-01

Dec-04

May-07

Nov-09

Apr-12

Date

Figure 24. Bottom hole pressure profile from Baturaja


Formation of Fika Field.

Wtihdrawal/Y1, MMSTB

Straight Line Material Balance Plot With Aquifer


60
50
40
30
20
10
0
0

Carbonate facies on Fika Field geologically


were divided by reef, platform, and breccia conglomerate. Based on core analysis, the carbonate
facies were divided into six lithofacies as justification on permeability prediction. TZA was
divided into three rock types which represent
flow unit as low quality, medium quality, and
high quality limestones.
All reservoir property models were built
taking into consideration the complexity of
rock and oil types, because the properties were
tied into facies and water saturation distribution
based on the relationship between J- function
and capillary pressure.
Reserve calculation from static model was
confirmed by comparing reserve analysis with
material balance analysis on production data of
about 26 MMSTB.
Further research into the complexity of rock
and oil types, mainly for reservoir modeling,
should be continued using reservoir simulation so that oil flow behaviour can be observed
clearly.

IJ
O

(p.Qd)/Y1, M.psi

Figure 25. Straight line material balance plot with aquifer


of Fika Field.

Table 9. Aquifer Propeties defined for Material Balance


Analysis for Baturaja Formation in Fika Field
Aquifer Properties

h. ft

95

k.mD

106

w. cP

0.24

. fraction

0.18

cw, 1/psi

3.26E-06

cf, 1/psi

5.72E-06

Bw. RB/STB
re. ft

. degree

The authors wish to thank PT Medco E&P


Indonesia and SKK MIGAS for their permission
to publish this paper. In addition, the authors
would like to thank the management of PT
Medco E&P Indonesia for their encouragement
and support.

35.000
180

Based on the distribution of reservoir properties, i.e. porosity and water saturation, the
value of OOIP was about 25.3 MMSTB, while
the value of OGIP was about 131.7 BCF.
Because the hydrocarbon in-place has been
already confirmed, this static model could be
used in reservoir simulation (initialization and
history matching).
96

Acknowledgements

References

Asquith, G. B., 2004. Basic well log analysis


for geologist. American Association of Petroleum Geologist. Methods in Exploration.
Tulsa, Oklahoma, 16, p.12-135.
Erawati, F.A., 2013. Utilization of Advance Seismic Interpretation for Estimation Reservoir
Hydrocarbon Distribution on Carbonate,
FIKA Field Study Case, South Sumatera
Basin, M.S. Thesis, University of Indonesia.

Reservoir Modeling of Carbonate on Fika Field:


The Challenge to Capture the Complexity of Rock and Oil Type (E.F. Adji et al.)

Sumatera. M.S. Thesis, University of Brunei


Darussalam.
LAPI ITB, 2011. Geochemical Study and Evaluation in South Sumatra Basin (Soka, Lagan,
Matra and Iliran Regions), Final Report.
Rashid, H., Sosrowidjojo, I. B., and Widiarto,
F. X., 1998. Musi Platform and Palembang
High: A New up Straight Look at The Petroleum System. Proceedings of Indonesian
Petroleum Association, 26th Annual Convention and Exhibition, Jakarta, Paper, IPA 98
- I - 107, 11pp.

IJ
O

Handayani, R.S.W., Setiawan, D., and Afandi,


T. 2008.Reservoir Characterization of Thin
Oil Column to Improve Development Drilling in a Carbonate Reservoir: Case Study
of Gunung Kembang Fields. Proceedings
of Indonesian Petroleum Association, 32nd
Annual Convention and Exhibition, Jakarta,
Paper, IPA 08-E-160, 15pp.
Caroline L.T. J, 2005. Lithofacies Characterization of Soka Field Based on Core Calibration on Image Data and Wireline Log for
better Prediction of Reservoir Properties of
the Baturaja Formation, Soka Field, South

97

Indonesian Journal on Geoscience Vol. 1 No. 2 August 2014: 99-107

INDONESIAN JOURNAL ON GEOSCIENCE


Geological Agency
Ministry of Energy and Mineral Resources
Journal homepage: hp://ijog.bgl.esdm.go.id
ISSN 2355-9314 (Print), e-ISSN 2355-9306 (Online)

A Drowning Sunda Shelf Model during Last Glacial Maximum (LGM)


and Holocene: A Review
Tubagus Solihuddin
Center for Research and Development of Marine and Coastal Resources
Jln. Pasir Putih 1, Ancol Timur, Jakarta 14430

Corresponding author: solihuddin@gmail.com


Manuscript received: April 14, 2014, revised: July 16, 2014, approved: August 15, 2014

IJ
O

Abstract - Rising sea levels since the Last Glacial Maximum (LGM), some ~20,000 years ago, has drowned the
Sunda Shelf and generated the complex coastal morphology as seen today. The pattern of drowning of the shelf will be
utilized to assess likely timing of shoreline displacements and the duration of shelf exposure during the postglacial sea
level rise. From existing sea level records around Sunda Shelf region, sea level curve was assembled to reconstruct
the shelf drowning events. A five stage drowning model is proposed, including 1) maximum exposure of the shelf at
approximately 20,500 years Before Present (y.B.P.), when sea level had fallen to about -118 m below present sea level
(bpl.), 2) melt water pulse (MWP) 1A at ~14,000 y.B.P. when sea level rose to about -80 m bpl., 3) melt water pulse
(MWP) 1B at ~11,500 y.B.P., when sea level was predicted around -50 m bpl., 4) Early-Holocene at ~9,700 y.B.P,
when sea level was predicted at about-30 m bpl, and 5) sea level high stand at ~4,000 y.B.P., when sea level jumped
to approx. +5 m above present sea level (apl.). This study shows that the sea level fluctuated by more than 120 m at
various times during LGM and Holocene. Also confirmed that sea level curve of Sunda Shelf seems to fit well when
combined with sea level curve from Barbados, although the comparison remains controversial until now due to the
considerable distinction of tectonic and hydro-isostatic settings.
Keywords: Last Glacial Maximum, sea-level changes, transgression, drowning shelf

Introduction

The Sunda Shelf is located in Southeast Asia


and it represents the second largest drowned
continental shelf in the world (Molengraaff and
Weber, 1921; Dickerson, 1941). It includes parts
of Indonesia, Malaysia, Singapore, Thailand,
Cambodia, Vietnam coast, and shallow seabed
of the South China Sea (Figure 1). During the
LGM, when sea levels are estimated -116 m below present sea level (bpl.), the Sunda Shelf was
widely exposed, forming a large land so-called
Sunda Land connecting the Greater Sunda
Islands of Kalimantan, Jawa, and Sumatra with
continental Asia (Geyh et al., 1979; Hesp et al.,
1998; Hanebuth et al., 2000, 2009).

The Sunda Shelf is also considered as a


tectonically stable continental shelf during the
Quaternary (Tjia and Liew, 1996) and categorized
as a far field location (far away from former
ice sheet region), providing the best example for
observing sea level history and paleo-shoreline
reconstruction. In such environment, the effects
of seafloor compaction, subsidence, and hydroisostatic (melt water release from the ice sheets)
compensation are negligible during the relatively
short time interval of thousands of years (Lambeck et al., 2002; Wong et al., 2003).
This study reviews some published sea level
observations then presenting a summary of the
Sunda Shelf drowning model. Moreover, it discusses some sea level records from different lo-

IJOG/JGI (Jurnal Geologi Indonesia) - Acredited by LIPI No. 547/AU2/P2MI-LIPI/06/2013, valid 21 June 2013 - 21 June 2016

99

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 99-107


3000 m

Thailand

2500 m

0m

-2500 m

Sunda Shelf
-5000 m

-7250 m

Kalimantan
Su
m
at
ra

500

1000

Kilometre

elf
Sh
ul
Sah

Figure 1. A map shows the Sunda Shelf region derived from 30 arc-seconds resolution bathymetric grid sourced from GEBCO. (after Molengraaff and Weber, 1921; Dickerson, 1941).

Reconstructing Sea-level History

Studies on sea level history around Sunda Shelf


have been carried out by Geyh et al., (1979), Tjia,
(1996), Hesp et al., 1998, and Hanebuth et al.,
(2000, 2009) to provide information on paleoshoreline, paleo-river, and paleo-bathymetry.
The most relatively recent studies (Hanebuth
et al., 2000, 2009) demonstrated an important

IJ
O

calities in time-scale LGM to Holocene (Table 1)


and improves detailed colour maps of Holocene
sea level transgression on the Sunda Shelf (Voris,
2000; Sathiamurthy and Voris, 2006) in terms of
map resolution. The analyses and data presented
in this paper provide an up to date overview of the
history of sea level and paleo-shoreline changes
around Sunda Shelf region since the LGM to
Holocene.

Table 1. Sea Level Observations from some Localities presenting LGM - Holocene Sea-level Records
Localities

Barbados

Proxies

Coral, mostly
A. Palmata

Time scale

Dating

LGM - midHolocene

14

C and U-series

References

Uplift 0.34 mm/year

Fairbanks (1989),
Peltier and Fairbanks (2006)

Slow tectonic subsidence


(0.15 mm/year), far-field
location, less affected by
hydro-isostatic

Bard et al. (1996)

Tahiti

Fossil coral

MWP-IA mid-Holocene

14

Huon
Peninsula,
PNG

Coral (Porites,
Acropora,
Montipora,
etc.)

Post glacial Mid-Holocene

AMS radiocarbon Rapidly uplifting area (LIg


1.760.05 mm/year, MidHolocene 2.16 0.44 mm/
year, far-field location

Chappell and
Polach (1991)

Bonaparte
Gulf, NW
Australia

Sediment core,
marine shell,
fauna

Pre and post


LGM

AMS radiocarbon Relatively tectonically stable,


the effects of hydro-isostatic
are small

Yokoyama et al.
(2001)

Scott
Reef, NW
Australia

Coral

Holocene

U-series

Collins et al.
(2011)

100

C and U-series
where available

Tectonic
setting

Tectonic subsidence (0.29


- 0.45 mm/ year), far-field
location, less affected by
hydro-isostatic

A Drowning Sunda Shelf model during Last Glacial Maximum (LGM) and Holocene: A Review (T. Solihuddin)

Materials and Methods


The ocean topography and land data covering
the Sunda Shelf were extracted from the General
Bathymetric Chart of the Ocean Grid (Gebco) 0.8
grid with a spatial resolution of 30 arc-seconds
of latitude and longitude (1 minute of latitude =
1.853 km at the equator). The bathymetric grid
has largely been generated from a database of
over 290 million bathymetric soundings with
interpolation between soundings guided by
satellite-derived gravity data. Land data are
largely based on the Shuttle Radar Topography
Mission (SRTM30) gridded digital elevation
model (see web page at: http://www.gebco.net/
data and products/gridded bathymetry data/; accessed July 2013).
The extracted elevation data (in x, y, and z
coordinates) were exported into points in ASCII
format within GridViewer package programme.
These point data were then used to generate a
Digital Elevation Model (DEM) using a Triangulated Irregular Network (TIN) method or TIN
DEM within the Global Mapper v11.0 toolkit.
For the purpose of a two-dimensional layout, a
Grid DEM was generated and presented within
MapInfo7.0 software. All figures presented on
this paper are originally created by the author
following the method discussed above.
The sea-level curve estimation in Sunda
Shelf as shown in Figure 2 is derived from some
previous studies conducted in several localities
such as Strait of Malacca (Geyh et al., 1979),
Singapore (Hesp et al., 1998), and former North
Sunda River and Mekong Delta (Hanebuth et
al., 2000, 2009). This sea-level curve was then
correlated with the present-day topography
and bathymetry of the Sundaland to generate
maps and approximate shoreline configuration
of the Sundaland during the latest Quaternary.
However, several assumptions were made in the
workflow of this study as follows: 1) The current
topography and bathymetry of the Sunda Shelf
are only an approximation and do not reflect past
condition precisely. 2) The sea floor compaction,
subsidence, and vertical crust displacement due to
sedimentation, scouring, and tectonic processes
are not taken into account.

IJ
O

recent dataset from a number of sediment cores


which were dated by AMS radiocarbon, providing
records extending from LGM to Holocene that
fill some of the late-Glacial gaps from Barbados
records. The Sunda Shelf region is believed
to have been tectonically stable during the
Pleistocene (Tjia and Liew, 1996) and considered
as a far-field site where tectonic correction and
hydro-isostatic compensation are negligible.
The stages of rising sea levels on the Sunda
Shelf between ~21,000 y. B.P. and ~4,200 y. B.P.
were reported by Hanebuth et al. (2000). It was
initiated by the terminal phase of LGM sea level
lowstand (approximately -116 m bpl.) at about
21,000 y. B.P. and followed by transgression,
rising sea level to approx. -56 m bpl. at ~11,000
y. B.P. Whilst Geyh et al. (1979), Tjia (1996),
and Hesp et al. (1998) described the sea level
highstand and its gradual fall to current levels
thereafter in the Mid to Late Holocene. The
summary is as follows. In the EarlyHolocene
between 10,000 and 6,000 y. B.P., the sea levels
rose significantly from -51 m bpl. to 0 m (present
level). Following this, it reached a peak in the
Mid-Holocene between 6,000 and 4,200 y. B.P.,
exhibiting sea level highstand from 0 m to +5m
apl. After that, the sea level fell gradually until
reaching modern sea level at about 1,000 y. B.P.
(Figure 2).
20

Geyh et al., (1979)


Hesp et al., (1995)
Hanebuth et al., (2000)
Hanebuth et al., (2009)
Regression line order 3

Relative Sea Level (m)

-20

Holocene
highstand

-40

-60

-80

Transgression
LGM lowstand

-100

-120

-140
0

5000

10000

15000

20000

25000

Time (14C calibrated years BP)

Figure 2. The best-fit sea level curve estimation of Sunda Shelf


from 21,000 to 1,000 y. B.P. derived from Geyh et al., (1979),
Hesp et al., (1998), and Hanebuth et al., (2000; 2009).

101

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 99-107

separated. Adding that, the present-day Jawa Sea,


which connects Kalimantan and Jawa, was largely
inundated, separating partly the two mainlands
(Figure 3c). However, the Greater Sundaland
(i.e. Kalimantan, Jawa, and Sumatra) were still
connected to the Malaya Peninsula.
At approx. 9,700 y. B.P., when sea level was
predicted around-30 m bpl., the Jawa Sea became
a significant sea. The present-day Sunda , Karimata , and Malacca Straits, the land bridges that
connect the Greater Sundaland, were initially
inundated, forming a narrow channel among the
islands (Figure 3d). The marine transgression
reached a peak in the Mid-Holocene at approx.
4,000 y. B.P., rising sea level to about +5 m apl.
and drowning some lowland areas in the mainland
(Figure 3e). Finally, the sea level fell gradually
returning to present-day level at approx. 1,000 y.
B.P. (Figure 3f).

Maps
The maps presented in this paper show a
summary of the gradual Sunda Shelf drowning
model which represents the predicted shorelines
and shelf exposures during LGM and Holocene.
Starting from -118 m depth contour, the drowning
model was gradually established on a vertical elevation of -80 m, -50 m, -30 m, +5m, and present
sea-level which every depth contour corresponds
to 14C calibrated years Before Present age. For
example, the current -118 m depth contour was
predicted as a shoreline at approx. 20,500 y.
B.P., while the current -50 m depth contour was
attributed to 11,500 y. B.P., etc. Topographic and
bathymetric contours are indicated by the change
in colour scheme as shown in the legend; however, the grey colour is also applied to the DEM
representing the exposed shelf. In addition, the
flowage of paleo-river of Sunda Shelf during
LGM is also presented with refers to the map
of paleo-river (Voris, 2000; Sathiamurthy and
Voris, 2006).

IJ
O

Paleo-rivers on the Sunda Shelf


There were four large river systems on the
Sunda Shelf that drained the Sundaland during the LGM; the Siam River, the North Sunda
River, East Sunda River, and the Malacca Strait
River systems (Voris, 2000) (Figure 4). The Siam
River system which today is called Chao Phraya
included the river system of east coast of Malaya
Peninsula (Sungai Endau, Sungai Pahang, Sungai
Terengganu, and Sungai Kelantan) and part of
the Southwest Vietnam coast. Sathiamurthy and
Voris (2006) demonstrated that Sumatras Sungai Kampar also joined the Siam River system
through the Singapore Strait and then ran north to
the Gulf of Thailand where the major Siam River
system situated and drained to the large expanse
of Sunda Shelf.
The North Sunda River system was considered
as the major Sunda Shelf River system (Molengraaff Rivers of Dickerson, 1941; Kuenen,
1950; Tjia, 1980) which drained north to the sea
northeast of Natuna Island. This system included
some tributaries of Central and South Sumatra
coast (Sungai Indragiri, Sungai Batanghari, and
Sungai Musi) and the large Kapuas River system
from Kalimantan.
The East Sunda River system drained to the
east across what is the present-day Jawa Sea before flowing east to the sea near Bali. This system

Results

The Drowning Sunda Shelf History


The history of the drowning Sunda Shelf was
initiated at approximately 20,500 y. B.P. when sea
level had fallen to around -118 mbpl. By this time,
the Sunda Shelf was largely exposed, forming
a massive lowland which connects present-day
mainlands in this region (Kalimantan, Jawa, Sumatra, and Malaya Peninsula) (Figure 3a). During melt water pulse (MWP)-1A, some ~14,000
y. B.P. (Fairbank, 1989), sea level rose rapidly
to approx. -80 m bpl., inundating Sunda Shelf
around the present-day Natuna Island. However,
the mainlands were still connected to each other
and the configuration of the exposed Sunda Shelf
remained very similar to the -118 m bpl. formation (Figure 3b).
Following that, the sea-level still experienced
a rapid rise and jumped to around -50 m bpl.
at about 11,500 y. B.P. (MWP-1B of Fairbank,
1989), exhibiting initial isolation of Natuna and
the Anambas Islands from the mainland. Thus,
the connections between Kalimantan and Malaya
Peninsula via South China Sea were initially

102

A Drowning Sunda Shelf model during Last Glacial Maximum (LGM) and Holocene: A Review (T. Solihuddin)

3000 m

3000 m

2500 m

2500 m

0m

0m

-2500 m

-2500 m

-5000 m

-5000 m

-7250 m

-7250 m
250 km

500 km

750 km

1250 km

0 km

250 km

500 km

3000 m

2500 m

2500 m

0m

IJ
O

0m

-2500 m

-2500 m

-5000 m

-5000 m

-7250 m

-7250 m

250 km

500 km

750 km

1250 km

0 km

250 km

500 km

750 km

1250 km

250 km

500 km

750 km

1250 km

3000 m

3000 m

2500 m

2500 m

0m

0m

-2500 m

-2500 m

-5000 m

-5000 m

-7250 m
0 km

1250 km

3000 m

0 km

750 km

0 km

-7250 m
250 km

500 km

750 km

1250 km

0 km

Figure 3. Shelf and sea level exposures at various ages. a). 20,500 y BP, sea level of - 118 m bpl. b). 14,000 y BP (MWP
1A), sea level of - 80 m bpl. c). 11,500 y BP (MWP 1B), sea level - 50 m bpl. d). 9,700 y BP, sea level of - 30 m bpl-predicted. e). 4,000 y BP, sea level +5 m apl. f). Present day sea level. Map derived from 30 arc-seconds resolution bathymetric
grid sourced from GEBCO.

103

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 99-107

Chao
Mekong

Ma

r
ve
Ri
da
N.
Su
n

er

sia

v
Ri

lay

it
tra

aS
cc
ala

Sunda Shelf

m
Su

Kalimantan

ra
at

N
0

250

500

Jawa

Kilometre

E. Sunda River

Hence, it is necessary to consider those mentioned factors when combining the data into a
single dataset and generating sea level curve on
these data. Also, when comparing these sea-level
records to Sunda Shelf data, those factors should
be taken into account to avoid bias in analysis
and interpretation.
Fairbanks (1989) and Peltier and Fairbanks
(2006) reported an important source of information for relative sea-level changes in Barbados
during the late stages of the LGM and the lateGlacial period. Using coral cores as a proxy and
AMS radiocarbon calibrated by Thermal Ionisation Mass Spectrometry (TIMS) dating methods,
the local relative sea level in Barbados stood
between -125 m bpl. at 21,000 y. B.P. and -15
m bpl. at 7,000 y. B.P. (Figure 5). The uplift rate
was 0.34 mm/year due to local tectonic setting.
Meanwhile, the relative sea-level change data
in Tahiti is from Bard et al., (1996) with supporting information on coral species given by
Montaggioni and Gerrard (1997). Using coral as
a proxy, radiocarbon dating yielded time scale
between MWP-1A and Mid-Holocene (Figure
5). Tahiti experienced slow tectonic subsidence
(0.15 mm/year) and was also characterized as a
far-field location.
The local relative sea-level changes were also
investigated from a rapidly uplifting area such as
Huon Peninsula, Papua New Guinea. The records
were obtained from a raised Holocene reef drill
core collected by Chappell and Pollach (1991).
AMS radiocarbon dating was applied to the
samples and uranium series (U-series) ages were
subsequently obtained from the same samples by
Edward et al. (1993), providing sea-level indicators from post-Glacial to Mid-Holocene (Figure
5). The uplift rate was reported 1.76 0.05 mm/
year in the last-Interglacial (LIG) and 2.16 0.44
mm/year in the Mid-Holocene. This region was
also considered as a far-field location.
Moreover, Yokoyama et al. (2001) discussed
the relative sea-level estimation from the NW
Australia Shelf. The information was obtained
from the sediment cores of Bonaparte Gulf which
were dated by AMS radiocarbon dating, providing
sea-level indicators corresponding to a late stage
of LGM (Figure 5). The region of NW Australia
was assumed to be relatively tectonically stable

Figure 4. A map shows paleo-rivers on the Sunda Shelf


during LGM (After Voris, 2000; Sathiamurthy and Voris,
2006).

IJ
O

included present-day rivers of north coast of Jawa,


the south coast of Kalimantan and the northern
portion of the east coast of Sumatra. Some smaller
rivers in SE Sumatra and the Seribu Islands area
of Jawa Sea ran south via the Sunda Strait to
enter the Indian Ocean (Umbgrove, 1949; van
Bemmelen, 1949).
The Straits of Malacca River system had
two drainages separated by a topographic height
between the Bernam and Kelang Rivers. One
drained NW to the Andaman Sea, including some
tributaries of this river system i.e. Sungai Simpang Kanan, Sungai Panai, Sungai Rokan, and
Sungai Siak of east coast of Sumatra and some
rivers from the west coast of Malaya Peninsula
i.e. Sungai Perak, Sungai Bernam, Sungai Muar,
and Sungai Lenek. Whilst the other drained SW
and eventually joined the North Sunda River.
Discussion
Comparison with Data from other Studies
Some studies have been carried out from different localities to obtain sea-level stands during
LGM and Holocene. The observations resulted in
varied conclusions depending mainly on covering
time periods, proxies, dating methods, isostatic
effects, and vertical tectonic land movement.

104

A Drowning Sunda Shelf model during Last Glacial Maximum (LGM) and Holocene: A Review (T. Solihuddin)

20

U/Th Barbados coral (Fairbanks, 1989)

10

15

20

25

35

14

-20

C Tahiti coral (Bard et al., 1996)

C Huon Peninsula coral (Chappell and


Polach, 1991)

-40

C Bonaparte Gulf sediment core


(Yokoyama et al., 2001)

14

-60

U/Th Scott Reef coral (Collins et al.,


2011)

-80
-100
-120
-140

Best-t-sea-level curve for Tahiti


Best-t-sea-level curve for Huon
Peninnsula
Best-t-sea-level curve for
Baonaparte Gulf
Best-t-sea-level curve for Scott
Reef

IJ
O

-160

Best-t-sea-level curve for Barbados

Relative sea-level (m)

14

Age (x1000 calibrated y BP)

Figure 5. Sea-level curve estimations from some localities presenting LGM - Holocene records.

and was considered as a far-field site where the


effects of hydro-isostatic are small.
The most relatively recent studies (Collins
et al., 2011) demonstrated an important recent
dataset from a number of coral cores in Scott
Reef, Northwest Australia, which were dated by
high resolution U-series dating. The data provided
Holocene sea-level records that characterized by
moderate rates of sea-level rise of 10 mm/year and
confirmed tectonic subsidence of 0.29 - 0.45 mm/
year (Figure 5). The region was also less affected
by hydro-isostatic due to far-field location.
Despite the fact that Sunda Shelf and NW
Australia region are proximal and considered to
be tectonically stable at least during Holocene and
less affected by glacio-isostatic adjustment due
to their far-field locations, the sea level curve of
Sunda Shelf during LGM and Holocene seems to
fit well when combined with sea level curves from
Barbados. This is also in accordance with Peltier
and Fairbanks (2006) who reported that the combination between Bonaparte Gulf, NW Australia,

and Sunda Shelf records did not match together


well and suggested that Sunda Shelf records fit
much better with the Barbados dataset. However,
the tectonic and hydro-isostatic settings of Sunda
Shelf and Barbados differ considerably and that
comparison remains controversial until now.
Sea-level High Stand during Mid-Holocene
and There after
Another issue that arises in the discussion of
the Sunda Shelf sea-level history is the controversy of the precise details of the Mid-Holocene
highstand. The dispute is likely because of the
variation on timing, glacio-isostatic adjustment,
and localised tectonics. For example, one appraisal of evidence from Malay-Thai Peninsula
(Tjia, 1996) revealed that the sea level highstand
peaked at +4 m at 6,000 y. B.P. and +5 m at 5,000
y. B.P. Whilst, an earlier survey in the Strait of
Malacca, between Port Dickson and Singapore
(Geyh et al., 1979) evidenced the highest dated
level at +2.5 m to about +5.8 m for the time inter105

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 99-107

Conclusions

has helped a lot in map making and figures, also


productive discussions during reviewing an early
draft of manuscript. The author also thanks to Idham Effendi, researcher in Indonesian Geological
Agency, for his review of the manuscript and help
of the submission. This paper is a contribution to
Centre for Research and Development of Marine
and Coastal Resources, Ministry of Marine Affairs and Fisheries.
References
Bard, E., Hamelin, B., Arnold, M., Montaggioni,
L., Cabioch, G., Faure, G., and Rougerie, F.,
1996. Deglacial sea-level record from Tahiti
corals and the timing of global meltwater discharge. Nature, 382(6588), p.241-244. doi:
p.10.1038/382241a0.
Chappell, J. and Polach, H., 1991.Post-glacial
sea-level rise from a coral record at Huon
Peninsula, Papua New Guinea. Nature,
349(6305), p.147-149.
Collins, L. B., Testa, V., Zhao, J., and Qu, D.,
2011. Holocene growth history and evolution
of the Scott Reef carbonate platform and coral
reef. Journal of the Royal Society of Western
Australia, 94(2), p.239-250.
Dickerson, R. E., 1941. Molengraaff River: a
drowned Pleistocene stream and other Asian
evidences bearing upon the lowering of sea
level during the Ice Age. In: Speiser, E. A.
(ed.), Proceedings Universityof Pennsylvania,
Bicentennial Conference, p.13-20. University
of Pennsylvania Press.
Edwards, R. L., Beck, J. W., Burr, G. S., Donahue,
D. J.,Chappell, J. M. A., Bloom, A. L., Druffel,
E. R. M., and Taylor, F. W., 1993. A Large
Drop in Atmospheric 14C/12C and Reduced
Melting in the Younger Dryas, Documented
with 230Th Ages of Corals. Science, 260
(5110), p.962-968.
Fairbanks, R. G., 1989. A 17,000-year glacioeustatic sea level record influence of glacial
melting rates on the Younger Dry as event
and deep ocean circulation. Nature, 342,
p.637-642.
Geyh, M.A., Kudrass, H.R., and Streif, H., 1979.
Sea Level Changes during the Late Pleisto-

val between 5,000 and 4,000 y. B.P. Furthermore,


review from two areas in Singapore, Sungai Nipah,
and Pulau Semakau (Hesp et al., 1998) concluded
that the peak of the sea level at about +3 m rather
than +5 m between 6,000 y. B.P. and 3,500 y. B.P.
Despite such discrepancies, there is a general
consensus (Geyh et al., 1979; Tjia, 1996; Hesp et
al., 1998) that the sea level highstand was attained
by ~6,000 y. B.P. or slightly earlier. By that time,
the sea level was around +3 m to +5 m above present sea-level then receded to its current datum for
the past ~1,000 years. The noticeable impact of sea
level rise through the transgression is that shoreline position changed markedly and was higher
and landward of present level during the high
stand (~6,000 - 4,000 y. B.P.). As sea levels fell
post-highstand, the shoreline prograded seaward,
forming numerous beach ridges in the sequence.

IJ
O

The Sunda Shelf provides suitable environments for sea level studies and has provided one
of the best examples of sea-level at the time of
the LGM. In particular, the continent is relatively
tectonically stable and lies far away from the former ice sheets, thus the effects of hydro-isostatic
adjustment are less and eustatic changes should
be well reflected in the data. The model reveals
that the drowning Sunda Shelf was initiated at
~20,500 y. B.P., when sea level had fallen to
about -118 m bpl. During sea-level transgression,
the Sunda Shelf experienced rapid sea-level rise,
inundating the shelf exposures until reaching
sea-level highstand at ~6,000 - 4,000 y. B.P.,
before finally returning to the present sea level
at approx. 1,000 y. B.P. LGM and Holocene sea
level changes are principally forced by the climate
change; with sea level fluctuating by more than
120 m at various times. The trigger for these
climate and sea level variations is believed to
relate to cyclic changes in the earths orbit and
solar radiation (the Milankovitch cycles) and the
insolation of the worlds atmosphere and oceans.
Acknowledgements
The author would like to thank Moataz Kordi,
fellow PhD students at Curtin University, who

106

A Drowning Sunda Shelf model during Last Glacial Maximum (LGM) and Holocene: A Review (T. Solihuddin)

submerged lakes on the Sunda Shelf. The


Natural History Journal of Chulalongkorn
University, Supplement 2, p.1-43.
Tjia, H.D., 1980. The Sunda shelf, Southeast
Asia. Zeitschrift fiir Geomorphologie N. F.,
24, p.405-427.
Tjia, H.D., 1996. Sea-level changes in the
tectonically stable Malay-Thai Peninsula.
Quaternary International, 31(0), p.95-101.
Tjia, H. D. and Liew, K.K., 1996.Changes in
tectonic stress field in northern Sunda Shelf
basins In: Hall, R. and Blundell, D. (Eds.),
Tectonic Evolution of Southeast Asia. Geological Society Special Publications, 106,
p.291-306.
Umbgrove, J.H.F., 1949 Structural History of
the East Indies, xi-63pp. Cambridge University Press, Cambridge.
Van Bemmelen, R.W., 1949. The Geology of
Indonesia, 1A. Government Printing, 732pp.
Voris, H.K., 2000. Maps of Pleistocene sea
levels in Southeast Asia: shorelines, river
systems and time durations. Journal of Biogeography, 27, p.1153-1167.
Wong, H.K., Haft, C., Paulsen, A.M., Ldmann,
T., Hbscher, C., and Geng, J., 2003. Late
Quaternary sedimentation and sea level
fluctuations on the northern Sunda Shelf,
southern South China Sea. In: Sidi, F.H.,
Nummedal, D., Imbert, P., Darman, H.,
Posamentier, H.W. (Eds.), Tropical Deltas
of Southeast Asia - Sedimentology, Stratigraphy, and Petroleum Geology: Society
Economical Palaeontologists Mineralogists
Special Publication, 76, p.200-234.
Yokoyama, Y., De Deckker, P., Lambeck, K.,
Johnston, P., and Fifield, L.K., 2001. Sealevel at the Last Glacial Maximum: evidence
from northwestern Australia to constrain ice
volumes for oxygen isotope stage 2. Palaeogeography, Palaeoclimatology, Palaeoecology, 165 (3-4), p.281-297.

IJ
O

cene and Holocene in the Strait of Malacca.


Nature, 278, p.441-443.
Hanebuth, T., Stattegger, K., and Grootes, P.M.,
2000.Rapid Flooding of the Sunda Shelf:
A Late-Glacial Sea-Level Record. Science,
288(5468), p.1033-1035.
Hanebuth, T.J.J., Stattegger, K., and Bojanowski, A., 2009. Termination of the Last Glacial
Maximum sea-level lowstand: the Sunda sealevel record revisited. In: Camoin, G., Droxler, A., Miller, K., and Fulthorpe, C. (Eds.),
Records of Quaternary sea-level changes:
Global and Planetary Change, 66, p.76-84.
Hesp, P.A., Hung, C.C., Hilton, M., Ming, C.L.,
and Turner, I.M., 1998.A first tentative Holocene sea-level curve for Singapore. Journal
of Coastal Research, 14, p.308-314.
Kuenen, P.H., 1950. Marine Geology. John
Wiley & Sons, Inc., New York, vii-568pp.
Lambeck, K., Yokoyama, Y., and Purcel, A.,
2002. Into and out of the Last Glacial
Maximum: sea-level change during Oxygen
Isotope Stages 3 and 2. Quaternary Science
Reviews, 21(1-3), p.343-360.
Molengraaff, G.A.F. and Weber, M., 1921. On
the relation between the Pleistocene glacial
period and the origin of the Sunda Sea (Javaand South China-Sea), and its influence on
the distribution of coral reefs and on the
land- and freshwater fauna. Proceedings
of the Section of Sciences, 23, p.395-439.
[English translation]
Montaggioni, L.F. and Gerrard, F.,1997. Response of reef coral communities to sealevel rise: a Holocene model from Mauritius
(Western Indian Ocean). Sedimentology, 44
(6), p.1053-1070.
Peltier, W.R. and Fairbanks, R.G., 2006. Global
glacial ice volume and Last Glacial Maximum duration from an extended Barbados
sea level record.Quaternary Science Reviews, 25 (23-24), p.3322-3337.
Sathiamurthy, E. and Voris, H.K., 2006. Maps
of Holocene sea level transgression and

107

Indonesian Journal on Geoscience Vol. 1 No. 2 August 2014: 109-119

INDONESIAN JOURNAL ON GEOSCIENCE


Geological Agency
Ministry of Energy and Mineral Resources
Journal homepage: hp://ijog.bgl.esdm.go.id
ISSN 2355-9314 (Print), e-ISSN 2355-9306 (Online)

Depositional Cycles of Muara Wahau Coals, Kutai Basin, East Kalimantan


Komang Anggayana1, Basuki Rahmad2, and Agus Haris Widayat1
Research Group of Earth Resource Exploration, Faculty of Mining and Petroleum Engineering, ITB
Jln. Ganesha No. 10, Bandung
2
Study Programme of Geology, Faculty of Mineral Technologies, UPN Veteran Yogyakarta
Jln. Swk 104 (Lingkar Utara), Condongcatur, Yogyakarta

Corresponding author: haris@mining.itb.ac.id; komang@mining.itb.ac.id; b_rahmad2004@yahoo.com


Manuscript received: April 1, 2014, revised: May 12, 2014, approved: August 19, 2014

IJ
O

Abstract - Fifteen samples were taken ply by ply from a 33 m thick drill core of Muara Wahau coal seams for interpretation of depositional environments. Generally, lithotype variation in the bottom part of the coal seams has a
lower frequency than in the upper part. Petrographical analysis was performed to determine the maceral composition,
groundwater index (GWI), and gelification index (GI). The samples from lower sections show much higher GWI-GI
values and lower variation frequency than from the upper section. This characteristic is interpreted as the result of
development of mesotrophic to ombrotrophic peats during the deposition of lower to upper parts of the section, respectively. During the development of the mesotrophic peat, water was more abundant and relatively stable in budget.
However, during the development of ombrotrophic peat, water was less abundant and relatively not stable in budget.
The latter is related to the water supply depending only on rain, resulted in the more sensitive water table in the ombrotrophic peat. The unstable water table is thought as the reason of higher variation frequency of lithotype, GWI,
GI, as well as maceral composition in the upper part of the core. Unstable water table would lead to moist condition
in the uppermost layer of the ombrotrophic peat, favoring fungi to grow. This is confirmed by the higher abundance
of sclerotinite maceral in samples from the upper part of the coal core.
Keywords: coal seam facies, Muara Wahau, Kutai Basin

Introduction

Muara Wahau coals are part of Early Miocene


Wahau Formation, Kutai Basin, East Kalimantan,
consisting of several thick coal seams. The thickness of each seam is up to 60 m. The coal has
not been disturbed by exploitation, as the site is
located in a very remote area.
Peat accumulation is controlled by at least two
key factors: subsidence rate creating accommodation space and organic matter (trees) supply rate
(Taylor et al., 1998). The interplay between the
two factors in equilibrium state brings through
peat accumulation. The longer equilibrium state
time is reached, the thicker peat would be accumulated. Another important aspect is the presence
of water which takes a role as media of nutrient

supply and deposited-organic matter preservation. Peat water may originate from groundwater,
rain water, and intrusion of sea water. The peat
water table is fluctuative and sensitive to season
or climate changes. The hydrological condition
in peats determined the vegetation and peat type
accumulated in a basin.
In the present case, the Muara Wahau coals
are interesting to investigate. The thick coal
seams can provide more detailed indications of
chemical and physical changes during peatification. With respect to hydrological condition,
the succession of early development of Muara
Wahau coals will be reconstructed and discussed
in detail. Petrographical analysis was carried out
to characterize the maceral composition of the
Muara Wahau coals.

IJOG/JGI (Jurnal Geologi Indonesia) - Acredited by LIPI No. 547/AU2/P2MI-LIPI/06/2013, valid 21 June 2013 - 21 June 2016

109

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 109-119

Methods

The studied area is located in Muara Wahau


Sub-regency, East Kutai Regency, East Kalimantan, Indonesia (Figure 1). The area is mostly
composed of Early Miocene Wahau Formation
according to Supriatna and Abidin (1995) (Figure
2). The formation was deposited in Upper Kutai
Basin (Calvert, 1999). The lower part of the formation consists of interbedded coralline and algal
limestones. The upper part of Wahau Formation
comprises interbedded tuff, claystones, quartz
sandstones, sandy claystones, and coal layers.
There are two major coal seams in the area (Figure 3), showing variation in the thickness from
15 to 62 m, with gentle dips that range from 8o
to 12o. The cropline distribution is controlled by
the presence of syncline in the area. Generally,
the coals are blackish brown, hard, brittle, dull
in luster, and contain resin and fossilized wood
in some parts.

The study was commenced by collecting representative samples from the coal field. Drilling
programme, which was conducted, penetrated
Seam 1 and Seam 2 in the coal field (Figure 4).
The selected coal samples were then taken directly ply-by-ply/ply sampling (Thomas, 2002)
from Seam 1 core. The samples were separated
based on the lithotype observed along the core,
then they were crushed and split for further laboratory analysis.
For maceral composition, fifteen samples
were investigated using Zeiss Axio Imager A2m
polarized microscope in the Laboratory of Earth
Resources Exploration, Faculty of Mining and
Petroleum Engineering, ITB. During the maceral
analysis, five hundred points with a minimum
distance of 0.2 mm between each point were
counted from the polished sections. The maceral
composition is stated as percent (%-volume).

le
Te

IJ
O

Geological Settings

au

ve

Ri

Sepinang

ive

nR

ah
W

Muara Wahau

Sangkulirang

KALIMANTAN

Tanah Merah

Tanjung Bengalun
Sangkinah

Muara Kaman

Tenggarong

SAMARINDA

ak

Adara
Pulung

Muara Payang

East Kalimantan

Penajam

100
Kilometres

116o E

Figure 1. Location map of the studied area.

110

1o S

Belimbing
Longikis

115 E

BALIKPAPAN

Studied Area
o

0o

tra
it

ass
ar
S

Adara

1 N

Tanah Grogot

117 E
o

118 E
o

QUATERNARY

SURFACE
DEPOSITS

AGE

VOLCANIC
ROCK

INTRUSION

PERIOD

Depositional Cycles of Muara Wahau Coals, Kutai Basin, East Kalimantan (K. Anggayana et al.)

FORMATION

DESCRIPTION
Alluvium: pebble, cobble, sand, mud
and plant remain

HOLOCENE

PLEISTOCENE
Metulang Vulcanic: andesite, basalt,
lava, tuff, aglomerat breccia and
laharic

PLIOCENE

Sintang intrusion: stock and dyke of


andesite and diorite
Upper Wahau Formation:
intercalation of tuff, clastone, quartz
sandstone, clayish sandstone, sandy
claystone, and lignite
Lower Wahau Formation:
interbedded coral and algae limestone

MIDDLE

OLIGOCENE

LATE

EARLY

PALEOCENE

EARLY

EOCENE

TERTIARY

MIOCENE

LATE

Marah Formation: intercalation of


marl, claytone, conglomrate, and
limestone

Metaphorphic rock and ultra basic


rock

116 42'E

116 44'E

116 46'E

116 48'E

116 50'E

1 16'N

1 16'N

IJ
O

Figure. 2. Regional stratigraphy of Upper Kutai Basin (Supriatna and Abidin, 1995).

N
1

Km

Co

al

al

Co

1 12'N

am

Se

am

Se

PMB-01-08

1 12'N

GT-03

1 14'N

1 14'N

GT-02

1o10'N

1o10'N

116 42'E
o

116 44'E
o

116 46'E

LEGEND :

Geological Map of Study Area

Wahau Fm

116 48'E

''

1 8'N

1 8'N

Tomw

Strike adn deep strata

Sintang andesite intrusion

Anticline, Syncline

Lineament

River

A
0

100
m

Coal Seam 1
Coal Seam 2

116 50'E
o

Sample location
Coal Cropline Seam A
Coal Cropline Seam B

B
0

100
m

Figure 3. Geological map and cropline of Muara Wahau coals (modified from Supriatna and Abidin, 1995).

111

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 109-119


-1

Natural Gamma
Ray
CPS

34

Depth
1m:100m

Coal

Thickness (m)

1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
11.0
12.0
13.0
14.0
15.0
16.0
17.0

14.00

18.0
19.0
20.0
21.0
22.0
23.0
24.0
25.0
26.0
27.0
28.0
29.0
30.0

Seam -1

31.0
32.0
33.0
34.0
35.0

33.30 m

36.0
37.0
38.0
39.0
40.0
41.0
42.0
43.0
44.0
45.0
46.0
47.0

47.30

48.0

50.0
51.0
52.0
53.0
54.0
55.0
56.0
57.0
58.0
59.0
60.0
61.0
62.0
63.0
64.0
65.0
66.0
67.0
68.0
69.0
70.0
71.0
72.0
73.0
74.0
75.0
76.0

61.30 - 62.30 (1.00)

GWI =

Humocollinite + Minerals
Humocollinite + Humodetrinite

IJ
O

77.0

49.0

Vitrinite (huminite) reflectance measurement was


also carried out in the Centre of Geological Resources, Geological Agency, Ministry of Energy
and Mineral Resources. Mean random vitrinite
reflectance measurements were performed on the
surface of vitrinite particles under oil immersion.
Fifty points of vitrinite reflectance were taken on
each sample. The identification and classification
of macerals in this study are based on ICCP System 1994 (ICCP, 1998, 2001, 2011), Suarez-Ruiz
and Crelling (2008), and Sykorova et al. (2011).
In order to investigate the hydrological regime
of the ancient peat, two indices were determined.
Groundwater index (GWI) based on Calder et al.
(1991) was used to indicate the ratio of strongly
gelified macerals (humocollinite), as well as the
mineral matter, to weakly gelified ones (humotellinite, humodetrinite). Mineral matter is considered
to be related to water carrying detritral material
to the former peat during flood (Crosdale, 1995).
.........................(1)

78.0
79.0
80.0
81.0
82.0
83.0
84.0
85.0
86.0
87.0
88.0
89.0
90.0
91.0
92.0
93.0
94.0
95.0
96.0
97.0
98.0
99.0

100.0
101.0
102.0
103.0
104.0
105.0
106.0
107.0
108.0
109.0
110.0
111.0

112.0
113.0
114.0
115.0
116.0

61.30 - 62.30 (1.00)


113.70

The second indicator used in this study is gelification index (GI) based on Diessel (1986). GI is
a tool to contrast gelified macerals to nongelified
ones as the indicator of peat wetness. When the
peat is flooded by water, vitrinite and geloinertinite tend to form. On the other hand, when the
water table is lower and the peat surface becomes
dryer or moist, teloinertinite and detroinertinite
will form in the peat surface due to oxidation
(Lamberson et al., 1991)

117.0
118.0
119.0

120.0

Seam-2

121.0
122.0
123.0
124.0
125.0

15.90 m

GWI =

Huminite + Macrinite
...(2)
Fusinite + Semifusinite + Sclerotinite + Inertedotrinite

126.0

129.0

129.60

130.0
131.0
132.0
133.0
134.0
135.0
136.0
137.0
138.0
139.0
140.0
141.0
142.0
143.0
144.0
145.0
146.0
147.0
148.0

146.20 - 148.20 (2.00)

149.0
150.0
151.0
152.0
153.0
154.0
155.0
156.0
157.0
158.0
159.0
160.0

Figure 4. Coal seam profiles (Seam 1 and Seam 2) from


density log interpretations of Borehole GT-02.

112

GWI and GI have different concepts. GWI


changes reflect the fluctuation of water table which
could happen in both high and low moor peat
types. In high moor, the water table would typically raise, following the development of rising
peat surface. GI assumes that gelification process
takes place in reduced water where the water table
is static regardless of peat surface development.
The GWI could be correlated with GI on the
development of peat as follows:
1. GWI < 0.5 and weak gelification (GI < 1)
reflect lower water table of a marsh (Diessel,

127.0
128.0

Depositional Cycles of Muara Wahau Coals, Kutai Basin, East Kalimantan (K. Anggayana et al.)

1986) and ombrotrophic peat type (Calder et


al., 1991).
2. GWI 0.5 - 1 and moderate gelification (GI ~
1) reflect higher water table of a fen? (Diessel, 1986) and mesotrophic peat type (Calder
et al., 1991).
3. GWI > 1 and strong gelification (GI > 1) reflect higher water table of a wet forest swamp
(Diessel, 1986) and rheotrophic peat type
(Calder et al., 1991). GI > 1 is characterized
by limited influx of clastics.
Result and Discussions

IJ
O

Coal Lithotype
Lithotype classification used in this study is
based on Stopes (1919) and Diessel (1965). Coal
from the Seam 1 of the Muara Wahau Formation
could be classified into three lithotypes, those are
banded coal (clarain), dull banded coal (clarain),
and dull coal (durain). Coal lithotype profile shows
cycle changes in the vertical section as presented
in Figure 5. Lithotype variation may indicate
changes of vegetation type composing coals
(Bustin et al., 1983) or changes in sedimentary
facies, especially related to water table conditions
during peat depositional process. Facies changes
in both vertical and lateral trends could control the
variation of maceral composition.

Humodetrinite (Figure 6b) shows fragmented


materials associated with inertinite, liptinite, or
minerals. Densinite appears as a mixture of fine
fragments of vitrinite. It is more homogenous than
attrinite. Humocollinite (Figure 6c) was found as
homogenous, rounded to oval bodies, and often
isolated one within desmocollinite.
Cutinite (Figure 6d) is mainly present as thin
continuous bands in association with vitrinite
maceral. Suberinite (Figure 6d) looks as cell walls
filled by other macerals, typically humocollinite.
Resinite (Figure 6e) appears as rounded, oval,
and unstructured material. Sporinite (Figure 6e)
is present only in a minor amount. It appears as
individual body with distinct cell walls and a
higher relief.
Sclerotinite shows rounded to oval forms and
has high reflectance (Figure 6f). This maceral is
present in all coal samples.
Mineralogical analayis shows that the coal
seam from the Seam 1 contains pyrite which is
present mostly as fine crystals within the dense
macerals (Figure 6f). The coal seam profiles

Coal Petrology
Microscopic analysis shows that huminite
reflectance of the Muara Wahau coals is 0.44
% on the average, suggesting brown coal
maceral nomenclature based on ICCP System
1994. The coals predominantly consist of
huminite macerals, with minor liptinite and
inertinite (Figure 6). Huminite maceral of the
coals comprises humotelinite, humodetrinite,
densinite, and humocollinite. Liptinite maceral
consists of cutinite, resinite, suberinite, and
sporinite. Inertinite maceral is dominated by
fusinite, semifusinite, and sclerotinite.
Humotelinite (Figure 6a) is mostly found as
thick layers in association with humodetrinite
and cutinite, grey to dark in colour, sometimes
forming lighter layers. This maceral may
originate from the lignin of high plants.

0m

10 m

Sample Number
15
14
13
12
11
10
9
8
7
6
5
4

20 m

3
2

Legend:
Dull Coal
Dull Banded Coal

30 m

Banded Coal

GT-02, Seam-1
Figure 5. Coal lithotype profile of seam 1 from Drillhole
GT-02.

113

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 109-119


0

100 mikron

100 mikron

Suberinite
Humocollinite
Cutinite
Humodetrinite
Sclerotinite

Sporinite

Humotellinite

a
0

100 mikron

Resinite

d
0

100 mikron

Sclerotinite

Cutinite
Sporinite

Humodetrinite

b
0

Fusinite

100 mikron

Humodetrinite

Resinite

100 mikron

Pyrite

Sclerotinite

IJ
O

Humocollinite

Sclerotinite

Humodetrinite
Humocollinite

Figure 6. Photomicrographs of macerals in the Muara Wahau coals (Seam 1). a. Sclerotinite associated with huminite; b.
Sclerotinite and fusinite associated with humodetrinite; c. Humodetrinite in association with round bodies of sclerotinite
and humocollinite; d. Suberinite, sporinite, resinite, and cutinite as individual bodies with distinct cell walls; e. Sporinite,
resinite, and cutinite as individual bodies with distinct cell walls; f. Huminite macerals in association with round bodies of
sclerotinite and pyrite mineral. a, b, c, and f in reflected light mode, while d and e in fluorescence mode.

obtain the mineralogial distribution and variation


in vertical sequence.
Coal Facies/Coal Deposition
Based on the maceral quantification, the
Muara Wahau coals consist of higher amount
of huminite (73.4 to 88%), lower proportion of
liptinite (0.6%-6.8%), and inertinite (5.8%-18%)
macerals (Table 1). The abundance of the macerals and minerals along the vertical coal profile are
shown in Figures 7a - d.
Figure 7a shows the variation of huminite
macerals along the coal profile. Humotelinite and
attrinite amount to about 9% and 8% on the average, respectively, and exhibit no vertical variation. Humodetrinite and humocollinite average

114

at 47% and 18%, respectively. Generally, humodetrinite abundance exhibits a decrease in depth,
the opposite condition with that of humocollinite.
In segment 1 (sample 1 - 4), the variation is low,
while in segment 2 (sample 5 - 15) the variation
is higher. The higher variation of humodetrinite
- humocollinite in segment 2 may reflect that the
groundwater in the ancient peat fluctuated more
intensively.
Figure 7b indicates that the abundance of
liptinite macerals generally decrease in depth.
Sporinite and cutinite only appear in some of segment 2 samples. Resinite is consistently present in
all samples, averaging 1.3%. Alginite is present
only in some samples of segment 2. Suberinite
appears in sample 3 - 15 with varying amounts.

Depositional Cycles of Muara Wahau Coals, Kutai Basin, East Kalimantan (K. Anggayana et al.)

Table 1. Maceral Composition of Seam 1 the Muara Wahau Coal


Sample Number
Maceral group

Maceral subgroup Maceral

HUMINITE
Humotelinite
Telocollinite
Humodetrinite

10

11

12

13

14

15

% Volume of each sample


83.6

81.4

78.6

82.6

86.6

88.0

80.4

73.4

87.8

83.2

75.0

77.4

86.0

81.4

86.8

12.6

11.0

10.6

10.0

10.6

12.2

8.6

5.0

12.6

10.0

2.8

6.0

8.6

7.4

7.6

12.6

11.0

10.6

10.0

10.6

12.2

8.6

5.0

12.6

10.0

2.8

6.0

8.6

7.4

7.6

52.2

45.0

51.8

49.0

39.4

52.4

53.4

57.6

48.4

56.2

67.4

64.4

63.6

61.2

64.2

Attrinite

8.6

6.0

6.6

7.6

5.2

8.4

8.8

9.2

6.6

11.4

5.0

3.4

8.8

7.2

13.8

Densinite

43.6

39.0

45.2

41.4

34.2

44.0

44.6

48.4

41.8

44.8

62.4

61.0

54.8

54.0

50.4

18.8

25.4

16.2

23.6

36.6

23.4

18.4

10.8

26.8

17.0

4.8

7.0

13.8

12.8

15.0

18.8

25.4

16.2

23.6

36.6

23.4

18.4

10.8

26.8

17.0

4.8

7.0

13.8

12.8

15.0

0.8

0.6

2.0

3.2

1.0

1.6

1.6

4.8

1.8

4.0

6.8

5.6

3.0

2.4

1.4

Sporinite

0.0

0.0

0.0

0.0

0.0

0.0

0.2

0.0

0.0

0.6

0.2

0.6

0.2

0.2

0.2

Cutinite

0.0

0.0

0.0

0.0

0.2

0.0

0.0

0.8

0.2

0.4

0.4

1.0

Resinite

0.8

0.6

0.6

1.2

0.8

1.0

0.6

2.4

0.6

2.4

3.4

1.4

1.4

1.6

0.4

Alginite

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.2

0.0

0.0

0.2

Suberinite

0.0

0.0

1.4

2.0

0.0

0.6

0.8

1.6

1.0

0.6

2.8

2.4

1.4

0.6

0.4

Humocollinite
Corpogelinite

INERTINITE

0.2

LIPTINITE

8.8

12.0

15.0

7.6

6.4

5.8

13.6

18.4

6.4

9.6

11.8

13.6

8.6

12.0

9.4

Fusinite

0.4

0.2

1.2

0.8

0.4

0.0

1.2

1.2

0.4

0.6

0.4

0.6

1.4

0.6

0.0

Semifusinite

0.8

1.8

3.0

2.8

0.8

0.2

2.8

2.6

0.2

1.4

1.6

1.6

0.4

1.2

2.0

Sclerotinite

5.8

7.6

7.4

3.4

3.8

4.8

5.6

9.8

5.6

6.0

7.2

8.8

5.8

7.6

5.2

Inertodetrinite

1.8

2.4

3.4

0.6

1.4

0.8

4.0

4.8

0.2

1.6

2.4

2.6

1.0

2.6

2.2

Macrinite

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.2

0.0

0.0

0.0

0.0

6.8

6.0

4.4

6.6

6.0

4.6

4.4

3.4

4.0

3.2

6.4

3.4

2.4

4.2

2.4

4.6

2.0

0.0

0.0

1.6

3.6

1.2

MINERAL MATTER
Oxide

1.8

0.4

2.0

0.6

1.6

1.8

IJ
O

Pyrite

0.6

Clay

1.0

1.2

0.6

2.8

2.4

0.6

1.6

4.4

5.6

2.4

0.4

2.0

1.6

1.8

1.0

1.2

0.6

2.8

2.4

0.6

1.6

1.2

TOTAL (%)

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

Rv. mean (%)

0.45

0.46

0.45

0.45

0.45

0.43

0.45

0.43

0.43

0.42

0.44

0.45

0.45

0.45

0.44

Resinite may originate from various plants.


Thus, this maceral is not a paleoenvironmentalindicative in the present case. Suberinite originates from cork tissues, plant components which
are similar to cuticles deriving cutinite. The suberinite source is easier to be destructed during
coalification, consequently, suberinite is rare and
infrequently found in some coals (Taylor et al.,
1998). In the Muara Wahau coal, suberinite occurs in almost entire vertical coal seam profiles,
except in the lower part (sample 1 and sample 2).
This may indicate that during the deposition of
sample 1 - 2, trees were firstly not present in the
peat. After that, the peat was starting to develop
trees, forming ombrogenous peat.
In Tertiary coals, sporinite originates generally from pollen of angiospermae. Thus,
the mechanism transport of pollen to the peat
swamp might be dominaned by wind (Esterle
and Ferm, 1994). The preservation of spore and
pollen is generally/mainly controlled by depositional environments. Taylor et al. (1998) for

example, revealed that the marine and brackish


environments would be more effective for spore
and polen preservation than freshwater environments. In addition, the preservation of spore and
pollen would be less efficient when exposed to
atmosphere. In the Muara Wahau coal, sporinite
is present only in segment 2 which might not be
related to a specific environment. The absence
of sporinite in segment 1 may suggest that during the development of the segment, there were
very limited spore/pollen producing plants as
the peat was still mesotrophic and dominated
by aquatic plants.
Inertinite macerals of the Muara Wahau coals
mainly consist of sclerotinite maceral. In vertical sequence, sclerotinite is the most abundant
maceral in the top and bottom parts of the seam
(Figure 7c). Sclerotinite was derived from fungi
dwelling in moist environments. The presence of
sclerotinite in all investigated samples suggests
that moist conditions intermittently occurred during the development of the peat.
115

116

50

49

48

47

46

45

44

43

42

41

40

39

38

37

36

35

34

33

32

31

30

29

28

27

26

25

24

23

22

21

20

19

8,8

12,6

11,0

10,6

10,0

10,6

12,2

10

6,0

6,6

7,6

5,2

8,6

8,4

6,6

9,2

12,6

8,6

5,0

8,8

11,4

3,4

7,2

13,8

10,0

2,8 5,0

6,0

8,6

7,4

7,6

40

50

60

18,4

70

25,4

16,2

23,6

10,8

18,8

23,4

23,4

26,8

17,0

4,8

80

13,8

2,0

1,6

2,8

0.8 %

0.6 %

2.0 %

3.2 %

2.4 %
3.0 %
5.6 %
6.8 %
4.0 %
1.8 %
4.8 %
1.6 %
1.6 %
1.0 %

1.4 %

Maceral Group Liptinite (%)

0,6

2,4

= Resinite

7,4

2,4

0,2

= Alginite

0,8

0,6

0,6

1,2

0,6

0,8

1,0

2,4

3,4

1,4

1,4

= Sporinite

50

49

48

47

46

45

44

43

42

41

40

39

38

37

36

35

34

33

32

31

30

29

0,4

1,0

0,6

= Cutinite

50

40

30

0,8

1,0

0,2 0,6

0,8

0,2 0,6

0,6

1,4

1,6

= Desmocollinite

100

28

27

0,6

0,2 0,4

26 0,2

25

24

23

22

21

20

0,2

0,2

0,2 0,2 0,4 0,2 0,4

= Corpogelinite

90

83.6 %

81.4 %

78.6 %

82.6 %

20

19

18

17

16

15

= Suberinite

12
11
10
9
8
7
6
5

14
13

15

Sample No.

50

40

30

20

0,6

0,4

0,6

0,6

1,4

50

49

48

47

46

45

44

43

42

41

40

39

38

37

36

35

34

33

32

31

30

29

28

27

26

25

24

1,4

1,8

0,8

1,2

0,8

1,2

1,2

0,4

1,8

1,2

1,6

2,0

23 0,6 0,1

22

21

20

19

18

17

16

15

14

13

2,8

3,0

4,8

2,8

2,6

3,0

5,6

5,8

5,8

5,2

3,4

1,4

0,8

6,0

1,2

7,6

0,1
5,6

8,8

0,6

7,4

1,0

1,6

9,8

2,2

10

2,4

2,4

2,6

4,0

12

2,6

3,4

14

4,8

= Sclerotinite

= Inertodetinite

16

= Macrinite

Maceral Group Inertinite (%)

= Semifusinite

1,8

= Fusinite

7,6

IJ
O

12
11
10
9
8
7
6
5

77.4 %
75.0 %
83.2 %
87.8 %
73.4 %
80.4 %
88.0 %
86.6 %

86.0 %

14
13

15

81.4 %

86.8 %

14

13

= Telocollinite

Maceral Group Huminite (%)

30

43,6

44,8

7,0

12,8

15,0

Deepness (meter)

= Densinite

20
a

39,0

45,2

41,4

44,0

44,0

41,8

44,6

48,4

62,4

61,0

54,8

54,0

50,4

Sample No.

18

20

8.8 %

12.0 %

15.0 %

7.6 %

12.0 %
8.6 %
13.6 %
11.6 %
9.6 %
6.4 %
18.4 %
13.4 %
5.8 %
6.4 %

9.4 %

12
11
10
9
8
7
6
5

14
13

15

Sample No.

Deepness (meter)

Deepness (meter)

Deepness (meter)

50

40

30

20

Figure 7. Comparison of maceral distribution in vertical section of the Seam-1, Muara Wahau coals a. huminite; b. liptinite; c. inertinite; d. minerals.

50

40

30

20

18

17

16

15

14

13

50

49

48

47

46

45

44

43

42

41

40

39

38

37

36

35

34

33

32

31

30

29

28

27

26

25

24

23

22

21

20

19

18

17

16

15

14

13

0,4

1,8

1,8

2,0

3,6

1,6

1,0

1,2

0,4

3,0

2,4

2,2

2,6

1,8

2,8

2,4

1,6

1,2

0,6

0,6

1,2

5,6

2,4

3,0

2,6

1,0

2,6

4
d

4,4

0,4

0,6

0,6

10

Mineral Matter (%)

2,0

= Clay

= Pyrite

2,0

3,6

= Oxides

12

6.8 %

6.0 %

4.4 %

6.6 %

4.2 %
2.4 %
3.4 %
6.4 %
3.2 %
4.0 %
3.4 %
4.4 %
4.6 %
6.0 %

2.4 %

14

12
11
10
9
8
7
6
5

14
13

15

Sample No.

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 109-119

Depositional Cycles of Muara Wahau Coals, Kutai Basin, East Kalimantan (K. Anggayana et al.)

Maceral variation and distribution in vertical


section of the coal profile, as mentioned above
clearly reveal that the peat developed under
the mesotrophic to ombrotrophic conditions.
Mesotrophic peat was mainly developed in
segment 1 of the coal profile indicated by the
presence of aquatic plants. On the other hand,
the ombrotrophic peat might occur during the
deposition of segment 2 dominated by trees with
less water influence.
Figure 8 shows the vertical variation of GWI,
GI, humocollinite, inertinite, sclerotinite, and
mineral matter profiles. Segment 1 (sample 1 - 4)
shows lower fluctuation profiles, while segment 2
(sample 5 - 15) indicates high fluctuation profiles.
The change of the fluctuation pattern in the vertical
coal profiles matches with that of coal lithotype.
Segmen 1 represents the development of
the mesotrophic peat type. In the peat, the

Gelication Index (GI)

Gelovitrinite

10

Mesotrophic
High Tide

15
14
13
12
20
11
10
9
9 25 8
7 6
6
5
5
30
4
4
15

15
14 13
12 1110

15
14
13
11 12
9 10
7 8
6
5
4

15

Depth (m)

Depth (m)

25

30

35

35

40

45

50

0.1

0.5

1.0

5.0

1.0

10.0

45

20.0

25

50

Gelication Index (GI)

30
35

40

20

Depth (m)

Depth (m)

20

5 10 15 20 25 30 35 40

% Gelovitrinite

25
30
35

10

15

20

25

30

10

50

Segment
2

Segment
1

3
2

45

1
Sclerotinite

% Inertinite

20

15
14
13 12
10 9 11
8 76
5
4

40

45
50

15

Mineral Matter

35

40

10

15
14
13
12
10 11
9 7
8
6
5
4

20

45
50

15

40

Sclerotinite

10

IJ
O

25

Inertinite

10

Depth (m)

Ombrotrophic
Low Tide

Depth (m)

Groundwater Index (GWI)


30

hydrological regime was influenced by both


groundwater and rain water. Such condition
would lead to the relatively more stable water
table. Segmen 2 represents the development of
the ombrotrophic peat type. The hydrological
regime in the ombrotrophic peat relied only
on rain water. This condition would promote
less stable water table, as rain water is very
sensitive to season and climate changes. Water
table fluctuation might occur faster resulting
in a more frequent moist condition in the peat
surface. This could allow fungi to grow. Thus,
consequently the coal in segment 2 contains
relatively more sclerotinite.
In order to investigate the relation between the
above parameters, some correlation plots were
made. Some of the plots are shown in Figure 9,
while the complete coefficients of determination
are listed in Table 2. The parameters are

1
2

10

Mineral Matter

Figure 8. Vertical variation of GWI, GI, humocollinite, inertinite, sclerotinite, and minerals.

16

y= 9.986x+4.589
2
R =0.371

14

Gelication Index (GI)

12
10

8
6
4
2
0

0.2

0.4
0.6
Groundwater Index (GWI)

0.8

Figure 9. GWI versus GI correlation.

117

Indonesian Journal on Geoscience, Vol. 1 No. 2 August 2014: 109-119


Table 2. Correlation of GWI, GI, Macerals, and Minerals
No

Parameter

Correlation

Coefficient of determination (R2)

GWI vs. GI

0,37

GWI vs. humocollinite

0,96

GWI vs. inertinite

0,29

GWI vs. sclerotinite

0,39

GWI vs. minerals

0,24

GI vs. humocollinite

0,49

GI vs. inertinite

0,89

GI vs. sclerotinite

0,62

GI vs. minerals

0,02

10

Humocollinite vs. minerals

0,13

variation in depositional cycles. The variation


of lithotype and maceral composition reveals
two segments showing different patterns along
the vertical profile. Segmen 1 (lower part of the
coal profile) shows depositional cycles related
to relatively stable water table, as interpreted by
the less frequent changes of sclerotinite, GWI,
and GI. Segmen 2 (upper part of the coal profile)
shows depositional cycles related to less stable
water table, as it shows more frequent changes
of the latter parameters. Based on the relatively
low GWI, GI, and the absence of correlation
between the two parameters and mineral matter
content, the Muara Wahau coal is interpreted to
be deposited in mesotrophic peat (segment 1)
which subsequently developed to ombrotrophic
peat (segmen 2).

IJ
O

typically positively correlated. GWI is positively


correlated with GI, as both indices are sensitive
to groundwater influence. Both indices are also
positively correlated with mineral and some
macerals which are well preserved in a wet
condition under water table. Several parameters
show negative correlation as shown by inertinite
and sclerotinite vs. GWI and GI. These macerals
developed under a moist condition, in parts where
sometimes covered by water.
A strong negative correlation occurs in
parameters GI vs. inertinite. It is normal, as the
two parameters require opposite depositional
environments. Gelification can only occur
below water table (reduction) condition, while
inertinite could only be derived in moist or
oxidation condition. The correlation between GI
and mineral matter is very weak. It is confirmed
that the mineral might be epigenetic, rather
than related to groundwater flow in the peat.
This also confirms that the peat development
was not initiated by eutrophic peat type, but
it was directly commenced by mesotrophic
and then ombrotrophic peats. Eutrophic peat
generally depends on groundwater flow for
nutrient supply. Coal which was deposited in
such peatland should have well correlation with
mineral matter content.

Acknowledgment

The authors are most grateful for permission of


data publication from PT. Bhakti Persada Energi.
Supports from Physics Laboratory of Centre of
Geological Resources (PSDG) for petrographical analysis are gratefully acknowledged.
References

Conclusions
The thick Muara coal seam has been investigated and is considered representative to indicate
118

Bustin, R.M., Cameron, A.R., Grieve, D.A., and


Kalkreuth, W.D., 1983. Coal Petrology - Its
Principles, Methods, and Applications. Short

Depositional Cycles of Muara Wahau Coals, Kutai Basin, East Kalimantan (K. Anggayana et al.)

ICCP (International Committee for Coal Petrology), 2001. The new inertinite classification
(ICCP system 1994). Fuel, 80, p.459-471.
ICCP (International Committee for Coal Petrology), 2011. Organic petrology, macerals,
microlithotypes, lithotypes, mineral, rank.
ICCP Training Course on DOM, Universidade do Porto, Portugal.
Lamberson, M.N., Bustin, R.M., and Kalkreuth,
W., 1991. Lithotype (maceral) composition
and variationas correlated with paleowetland environments, Gates Formation,
northeastern British Columbia. International
Journal of Coal Geology, 18, p.87-124.
Stopes, M.C., 1919. On the four visible ingredients in banded bituminous coals. Proceedings of Royal Society, 90B, p.470-487.
Surez-Ruiz, I. and Crelling, J.C. (eds.), 2008.
Applied Coal Petrology. The Role of Petrology in Coal Utilization. Elsevier Ltd, 388pp.
Supriatna, S. and Abidin, Z.A., 1995. Geological Map of Muara Wahau Sheet, Scale
1:250.000. Geological Research and Development Centre, Bandung.
Sykorova, I., Pickel, W., Christanis, K., Wolf,
M., Taylor, G.H., and Flores, D., 2005. Classification of huminite ICCP System 1994.
International Journal of Coal Geology, 62,
p.85-106
Taylor, G.H., Teichmuller, M., Davis, A., Diessel, C.F.K., Littke, R., and Robert, P., 1998.
Organic Petrology. Gebruder Borntraeger.
Berlin . Stuttgart. 704pp.
Thomas, L., 2002. Coal Geology. John Wiley
& Sons Ltd, England, 384pp.

IJ
O

Course Notes, 3. Geological Association of


Canada, Victoria.
Calder, J.H., Gibling, M.R., and Mukhopadhay,
P.K., 1991. Peat formation in a Westphalian
B piedmont setting, Cumberland basin, Nova
Scotia: implications for the maceral-based
interpretation of rheotrophic and raised paleomires. Bulletin de la Societe Geologique
France, 162, p.283-298.
Calvert, S.J., 1999. The Cenozoic Evolution of
the Larlang and Karama Basin, Sulawesi,
Proceedings of IPA 28th Annual Convention,
p.97- 115.
Crosdale, P.J., 1995. Coal facies studies in
Australia. International Journal of Coal
Geology, 58, p.125-130.
Diessel, C.F.K., 1965. Correlation of macro- and
micropetrography of some New South Wales
coals. In: Woodcock, J.T., Madigan, R.T., and
Thomas, R.G. (eds.), Proceedings-General,
8th Commonwealth Mineral and Metallurgy
Congress, 6, p.669-677.
Diessel, C.F.K., 1986. On the correlation between coal facies and depositional environment, Advances in the Study of the Sydney
Basin. Proceedings of the 20th SymposiumThe University Newcastle-Department of
Geology, Publication, 246, p.19-22.
Esterle, J.S. and Ferm, J.C.,1994. Spatial variability in modern tropical peat deposits from
Sarawak, Malaysia, and Sumatra, Indonesia:
analogues for coal. International Journal of
Coal Geology, 26, p.1-41.
ICCP (International Committee for Coal Petrology), 1998. The new vitrinite classification
(ICCP System 1994). Fuel, 77, p.349-358.

119

You might also like