PROCEEDINGS

JOINT CONVENTION MALANG 2017, HAGI – IAGI – IAFMI- IATMI (JCM 2017)
Ijen Suites Hotel, Malang, September 25 – 28, 2017

A Geochemical Approach To Evaluate Low Quality Volcaniclastic Reservoir

Dipo Caesario1, Wahyu Budhi K 1, Arif Swastika 1, Nanda Natasia 2, Galih Regi R 3, Wahyu Nainggolan 3, Yan
Wijaya 4,
1
PT. FERG Geosains Indonesia, 2 Fakultas Teknik Geologi Universitas Padjadjaran, 3 Samudra Energi-BWP
Meruap, 4 PT. Pertamina EP

Abstract Benakat Fm. (Figure. 1), using GLT (Geochemical Logging

Meruap KSO is a prolific oil block located in Jambi Sub- Tools) to determine the quality of reservoir.
basin, South Sumatra Basin. Volcanic activity in the
Miocene supplies the volcanic material to Air Benakat
Formation as the primary reservoir. There are two types of
lithology that strongly influenced by volcanic material,
namely pyroclastic-fine tuff and volcaniclastic-
conglomerate to volcaniclastic-sandstone. These consist of
volcanic glass (rhyolite composition) with low density,
high potassium content, altered volcanic glass products
(zeolite, calcite, siderite) contribute to give low grain
density, and demonstrate anomalous conventional log
reading (high gamma ray and neutron-density crossover). A
new method is proposed to solve this challenging reservoir
case through geochemical analysis by using Geochemical
Logging Tools (GLT).

The first step is core evaluation using thin section, XRD

and XRF. Chemical and physical parameters are
determined from minerals configure the matrix in
petrophysical model, categorized as QFM radioactive and
non-radioactive, clay minerals, and carbonate based on Figure 1: Paleogeography of Air Benakat Fm. Show there
spectrolith model (Herron ,1988). RCAL and XRF result is some Tuffaceous Sediment influence (Ginger
give an average grain density of 2.49 g/cc and Ca 0.24 %, & Fielding, 2005)
Fe 1.65 %, K 3.38%, Na 3.90%, Si 34.21% and Ti 0.1%
elements content, classified as modified Rhyolithic rock
fragment. GLT data combine with spectrolith model and Data and Method
Th-K-U ratio plot are used for predicting grain total Petrophysical Evaluation Methods (Figure-2) were
density, porosity, bound water, CEC, m, permeability performed on three wells which have GLT data (Elemental
lambda, and Swirr. Capture Spectroscopy & Spectral Gamma Ray) which are
AB-1, AB-2, & AB-3.
Thus, we interpret Air Benakat Formation in Meruap KSO
has at least five reservoir types classified as pyroclastic,
volcanogenic, and volcanoclastic. The pyroclastic group
has lower quality than the others with low resistivity
condition caused by conductive mineral.

Introduction
Meruap block is one of the areas of oil exploration and
production of 254.94 km2 as part of a Jambi Sub-Basin,
South Sumatra Basin (back arc basin) that confined by
Barisan Ridge Mountain on South West area.
Administratively, Meruap block is located in the
Sorolangun District, Jambi Province. Currently, the
Figure 2: Petrophysics for volcaniclastics reservoir using
exploration and production activities in this area is
managed by a joint operation of KSO Pertamina EP- Geochemical Logging Tools.
Samudra Energy-BWP Meruap. The area consists of two
main structures that are restricted by a series of West-East Core and Petrography Evaluation
trending inverted fault. This paper delivers all necessary From core observation, there are two kind of volcanic
stages of petrophysical evaluation for volcaniclastics of Air influence’s reservoir in this area, fine ash tuff with whiteish
PROCEEDINGS
JOINT CONVENTION MALANG 2017, HAGI – IAGI – IAFMI- IATMI (JCM 2017)
Ijen Suites Hotel, Malang, September 25 – 28, 2017

color (Figure 3.a) and greyish volcaniclastic sandstone with series) with rhyolite composition with 2.49 density (Table.
coarse sand to conglomeratic grain size (Figure 3.b). 1).
Routine core analysis shows that fine ash tuff has lower Table 1. Dry weight average of oxide and element from 4
permeability and grain density than volcaniclastics even sample fine ash tuff and their density average.
they have higher porosity. This indication suggests there
are different pore type and matrix for this two type
lithologies. From SEM and petrography data, tuff has
several microporosity indication with higher volcanic glass
content, and in volcaniclastic sandstone, volcanic glass
presents as fragment and also shows dissolution porosity.

Petrophysics Model For Volcaniclastic

Because of high-K content & low grain density of tuff and
tuff fragment (volcaniclastics sandstone fragment), new
petrophysics methods are proposed (Figure. 4). First, a new
model of petrophysics is created to accommodate tuff
content as matrix based on SEM, petrography, and XRD
analysis (Figure 4.a). Then, from ECS log & XRF analysis,
the composition of rock are obtained using spectrolith
model from Herron, 1988 (Figure 4.b).
(a)

(a)

(b)

(b)
(c)
Figure 3: (a) Ash Tuff, (b) coarse sand to conglomeratic Figure 3: (a) Petrophysic model for high-K content of
volcaniclastic sandstone, and (c) Geochemistry volcaniclastics in Meruap, (b) Matrix
composition in tuff (Le Maitre et.al. opcit composition determination modified from
Rollinson 1993). Herron 1988.

XRD analysis failed to distinguish volcanic glass as a Volume Radioactive & Non-Radioactive Minerals
matrix composition in tuff, they even read them as quartz, The presence of high-K tuff & low-density volcanic glass
but in XRF analysis, (alkali total (K2O+Na2O) vs silica in ash tuff and fragment of volcaniclastic sandstone could
total (SiO2) and K2O vs SiO2 crossplot) provided the make ambigous interpretation in Vshale GR determination
general composition of this tuff as a high-K (calc alkali to distinguish tuff, shale, and matrix. (tuff has higher GR
reading than claystone) (Figure 3.a & 3.b). Moreover, from
PROCEEDINGS
JOINT CONVENTION MALANG 2017, HAGI – IAGI – IAFMI- IATMI (JCM 2017)
Ijen Suites Hotel, Malang, September 25 – 28, 2017

spectrolith model Vclay (ClaySpectroplith) could obtained, but Wi : Dry volume mineral i
the clay mineral composition should present to provide a 𝜌i : Density of mineral i
better porosity determination, so Spectral Gamma Ray, The quality of mineral composition is controlled by
Th/K & Th/U analysis from Macfarlane,1989 is used. Th/K correlation of 𝜌matrix and 𝜌 grain density from routine
ratio is used to interpret clay mineralogy for each depth core analysis data (cc>0.75) (Figure 6).
(Figure 5.a). Tuff has lower Th/K ratio (<2ppm/%) and
from petrography, SEM, and XRD observation, some of
samples have altered to other minerals and clay minerals,
such as chlorite, illite, smectite, kaolinite, zeolite, siderite
& kalsit. To be simplified, mineral clay composition is
divided into two groups, illite zone (Th/K <3,5 ppm/%) and
smectite, chlorite, kaolinite zone (Th/K >3,5 ppm/%).

Figure 6: The result of mineral volume and grain density

prediction using GLT.

From the composition of total clay, QFMRAD, QFMNR, and

geochemistry composition (Na2O+K2O vs SiO2 crossplot),
a new lithology determination for this field is obtained.
Figure 5: Th/K analysis to distinguish clay minerals. They are pyroclastic (rhyolite composition & QFMRAD>0),
volcanogenic (QFMRAD=0), and volcaniclastic (non
rhyolite composition & QFMRAD>0). Then, a lithofacies
To distinguish volume radioactive & non-radioactive classification is achieved from combination of Schmid
minerals, the combination of QFMspectrolith and Vsh Gamma (1981) and Guilbert op.cit. William et.al (1982), the results
Ray are used. In a depth data which has XRD data and no are Shale (volcaniclastic & volcanogenic with Vclay>0,5),
volcanic content, QFMspectrolith = Vsh Gamma Ray, other Sandstone (rhyolite composition but without QFMRAD
depth data that has a difference in volume, QFMradioactive composition), Tuffaceous sandstone (volcaniclastics &
(QFMRAD) and QFMnonradioactive (QFMNR) are obtained. For pyroclastic with QFMRAD/QFMNR<25%), Pyriclastic-tuffit
VRAD, the compositition are rhyolite tuff, biotite, & zeolite (volcaniclastic & pyroclastics with
and for VNR are quartz, albite, pyrite. Then from carbonate 75%>QFMRAD/QFMNR>25%), and tuff pyroclastics
spectrolith, composition of calcite and siderite are obtained. (volcaniclastic & pyroclastics with QFMRAD/QFMNR>75%)
(figure 7.b).
To get a volume for each minerals, a transformation from
dry weight element to volume of dry mineral from Zhou, Porosity, Permeability Lambda (Kλ), & Irreducible
et.al. 2014 (1) & (2) are used and grains density prediction Water Saturation (SwIrr)
is gained from Bronnec et.al, 2006 (3). Neutron-density porosity method is used to get effective
porosity and total porosity with some modifications by
............(1) using different matrix density (𝜌m𝑎𝑡𝑟𝑖x prediction) for each
depth, clay volume from Clayspectrolith, and dry clay density
by averaging the composition of mineral models. Then
............(2) Sonic porosity is used to obtain m (cemetation factor),
using rasmus formula. Then Kλ and SwIrr are obtained using
formula (4) and (5) (from Tiwari, 2008 and Barson et. Al,
............(3)
2005) (figure 7.a).

Remarks: ............(4)
𝐷𝑊𝑋𝑋_𝐴 : Parameter respons dry weight for each minerals
𝐴𝑟𝑥𝑥 : Atomic number element
𝑁𝑥𝑥 : Sum of atomic element XX in mineral A ............(5)
𝑀𝑟𝐴 : Sum of atomic mass molecule in mineral A Remarks:
𝜌m𝑎𝑡𝑟𝑖x : Matrix density prediction z : Optimation factor
Ø : Total porosity
PROCEEDINGS
JOINT CONVENTION MALANG 2017, HAGI – IAGI – IAFMI- IATMI (JCM 2017)
Ijen Suites Hotel, Malang, September 25 – 28, 2017

Φ : Efective porosity increasing of depth, whereas porosity and permeability are

m : Cementation factor declined. If the curve is overlaid with PEF and resistivity
Soi : Ration of grains surface area to porosity volume log, it shows low resistivity effect caused by high siderite
(clay=6, QFM=0.22, carbonate=2, pyrite=0,1) and pyrite contents.
Mi : Volume of minerals i
ρm : Density of matrix

Figure 8: K vs CIA crossplot show quality of pyroclastic

Conclusion
High-K rhyolite tuff or fragment tuff components in
volcaniclastic affects the wireline log reading in gamma ray
and density log. Thus, the usage of GRmatrix, GRShale, or
Density matrix parameter is highly discouraged for
petrophysical analysis As the depth is increasing, it results
in tuff alteration followed by bad quality reservoir (in terms
of porosity, permeability, and Swirr values), besides the low
resistivity indication.

References

Bronnec R., 2005, GLT Processing on ARCO #16, Elnusa.

Fisher, R. V. and Schmincke, H.U., 1984, Pyroclastic
(a) Rock, Springer-Verlag, Berlin,
Ginger, D. and K. Fielding., 2005, The Petroleum Systems
and Future Potential of the South Sumatra basin,
Proceedings 30th Annual Convention Indonesian
Petroleum.
Herron, M. M., and Susan, L. H., 1997, Log Interpretation
Parameters Determined from Chemistry,
Mineralogy and Nuclear Forward Modeling,
Schlumberger-Doll Research, Ridgefield.
Rollinson H., 1992, Using Geochemical Data: Evaluation,
Presentation, Interpretation, Pearson Prentice
(b) Hall, Inggris.
Figure 7: (a) Porosity, Permeability lambda, and Swirr Schmid, R., 1981, Descriptive Nomenclature and
result, and (b) quality for each lithofacies. Classification of Pyroclastic Deposits and
Fragment: Recommendation of the IUGS
Subcommission on Systematics of Igneous Rock,
Geochemistry data for low resistivity analysis Geology, Zürich.
The reservoir quality of volcaniclastic and pyroclastics in Tiwari, R. R., 2008, New Generation Permeability Logs, 7th
this field is analyzed through geochemical approach, International Conference & Expisotion on
affected by weathering and alteration which represented as Petroleum Geophysics, Hyberabad.
K (weathered-crust) (6) and CIA coefficient (Chemical Zou, C., 2013, Volcanic Reservoirs in Petroleum
Index of Alteration) (7) (Zou, 2013). Exploration, Petroleum Industry, Press Elsevier
Inc., China.
...(6)
...(7)

On the croosplot of CIA vs K (figure 8), there is an

increased trend of weathering (CIA and K), aligned with