See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/344447970

Exploration Risk Assessment of the Mt. Talang Geothermal System, Sumatra,

Indonesia

Preprint · October 2020

CITATIONS READS

0 166

5 authors, including:

Rizal Abiyudo William B Cumming

MedcoEnergi Cumming Geoscience
10 PUBLICATIONS   1 CITATION    44 PUBLICATIONS   463 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

UNEP ARGeo Technical Advisor View project

Tolhuaca View project

All content following this page was uploaded by Rizal Abiyudo on 03 October 2020.

The user has requested enhancement of the downloaded file.

GRC Transactions, Vol. 44, 2020

Exploration Risk Assessment of the Mt. Talang Geothermal

System, Sumatra, Indonesia

Kusuma Putra1, Rizal Abiyudo 2, Munazyi2, W. Cumming3 and E. Easley4

1HitayEnergy Holdings
2University
of Indonesia
3Cumming Geoscience
4Thermochem

Keywords
Vapor Core, MT Resistivity, Magmatic, Geothermal Exploration Pitfalls, Geothermal Risk,
Mt. Talang, Indonesia

ABSTRACT

As lower risk prospects in Indonesia are developed, the next phase of high temperature geothermal
exploration will include an increasing focus on exploration for neutral reservoirs adjacent to vapor
core volcanoes, like Mt. Talang in Western Sumatra. The existence of a neutral geothermal
reservoir located below the flank Mt. Talang volcano, that very likely has a vapor/acid core, can
be discriminated by the neutral and the vapor core systems using water and gas geochemistry from
springs and fumaroles, magnetotelluric (MT) resistivity imaging, geology and structural data sets.
Mt. Talang is a predominantly andesite volcano within a trans-tensional basin of the Great Sumatra
Fault Zone. The volcanic cone is constructed on an ancestral volcano exposed to the south and
collapsed to the north. A vapor core has been interpreted from the gas chemistry of summit
fumaroles, Mt. Talang’s thirteen small magmatic eruptions and surface acid alteration. The
preferred conceptual model for Mt. Talang is analogous to a relatively small Mt. Apo, with a
neutral, mature, liquid system isolated from a magmatic vapor core by a low permeability shell of
clay and anhydrite alteration. Although this strategy for exploring vapor core volcanoes avoids
many pitfalls by building conceptual models and testing them against analogs in order to
discriminate neutral and vapor core zones, the risk of targeting wells adjacent to vapor core systems
can be further mitigated by using lower cost slim hole drilling.

1. Introduction
The Mt. Talang geothermal prospect is one of several prospects explored for geothermal power
development by Hitay Energy Holdings. The higher risk targets include geothermal systems that
have limited indicative thermal manifestation geochemistry and geothermal systems associated
with active volcanoes likely to host a vapor core analogous to the Alto Peak field in the Philippines
(Reyes et al., 1993) and the Tangkuban Perahu, Ijen, Bromo, Dempo, Galunggung, Arjuno-
 Putra et al.

Welirang and Slamet prospects in Indonesia (Hochstein and Sudarman, 2015; Abiyudo et al.,
2016). The Mt. Talang Geothermal System (Mt. Talang) is likely categorized a vapor core paired
with a neutral geothermal reservoir.
The term ‘vapor core’ was introduced by Reyes et al. (1993), to describe the Alto Peak system,
consisting of a chimney-like acidic vapor-dominated zone with a diameter of about 1 km, height
of 2-3 km, linking deep, and magmatic high-temperature zones. Allis (2000) suggests that, on rare
occasions with optimal circumstances, these systems might evolve into a neutral steam-dominated
reservoir, like the Darajat and Kamojang fields in Java. Other models for magmatic systems (e.g.,
Ellis and Mahon, 1977) consider the case where meteoric water floods a vapor core creating a hot
acid liquid reservoir at a higher pressure that may outflow at temperature >250°C. This gradually
becomes more neutral as surrounding rock is altered. Ramos-Candelaria et al. (1995) proposed the
presence of a vapor core in the conceptual models of several Philippine geothermal fields.
Hochstein and Sudarman (2015) point out that the chloride-sulfate hot springs on the flanks of
volcanoes hosting vapor core geothermal systems in Indonesia likely result from acid condensate
mixing with meteoric water near the vent areas that flow at shallow depth down the flanks of the
volcano, where fluids are progressively neutralized by rock alteration and dilution with
groundwater. Abiyudo et al. (2016) further points out that the alteration associated with shallow
advective flow down the flanks is supported by the general effusion of gas from volcanoes that
erupt repetitively, creating the extensive shallow smectite clay alteration layer that blankets many
of these vapor core volcanoes.
Based on vapor core conceptual models associated with exploration and development case
histories in Indonesia and Philippines, Abiyudo et al. (2016) proposed three types of systems that
a cost-effective geothermal exploration strategy should be designed to discriminate at volcanoes
likely to host a vapor core, like Mt. Talang. The three types and analogs include: 1) paired systems
with a neutral liquid zone adjacent to an acid magmatic vapor core like Mt. Apo in the Philippines
(Reyes et al., 1993) and Miravalles in Costa Rica (Gherardi et al., 2002; Marini et al., 2003); 2)
paired systems with a vapor dominated neutral zone adjacent to an acid magmatic zone like Patuha
(Layman and Sumarinda, 2003; Ashat et al., 2019) or Karaha-Bodas (Allis et al., 2000); and 3)
single vapor core systems with no significant associated neutral reservoir like Alto Peak (Reyes et
al., 1993). More recent geothermal field efforts have further supported this approach, such as the
Sorik Marapi field development (Licup et al, 2017; Sarmiento et al., 2017).

2. Geothermal Conceptual Models with a Vapor Core

Figure 1 illustrates conceptual cross-sections of a single vapor core volcano and a vapor core
volcano paired with a neutral liquid-dominated geothermal reservoir. In both cases, gases
including CO2, H2S, SO2 and HCl from the magmatic vapor core dissolve and oxidize in meteoric
and condensate water at shallow depth in the vent areas near the peak. The resulting shallow acidic
chloride-sulfate water alters overlying and adjacent rocks to low permeability kaolinite and alunite
clay and other indicative minerals, primarily sulfur, opal, and pyrite (Reyes, 1990), whereas
smectite forms as the acid water neutralizes through water-rock interaction and/or dilution. The
resulting clay-rich alteration zone is usually imaged by MT surveys as a conductive layer following
topography down the flank of the volcano. The low resistivity layer associated with the outflow
from the vapor core has sometimes been mistaken for a clay cap covering a neutral geothermal
reservoir. Resistivity is reduced by the near-surface clay alteration, mainly associated with the
 Putra et al.

smectite clays characteristic of neutral to moderately acidic conditions at <180°C, typically below
10 ohm-m in andesite terrain, whereas both unaltered rock and alteration at >230°C will be higher
resistivity (Usher, 2000).
In Figure 1b, the low permeability smectite-kaolinite clay that perches the near-surface chloride-
sulfate outflow also acts as a cap at lower elevation over an underlying neutral reservoir. The acid
vapor core chimney is isolated from the neutral mature liquid system by a low permeability shell
of anhydrite and clay alteration that encloses the deep magmatic intrusion and shallow volcanic
neck. As in the case of a conventional liquid geothermal reservoir, the hot upflow is diverted by
the smectite clay cap that channels the buoyant hot water along an outflow path (Hochstein and
Sudarman, 2015; Abiyudo et al., 2016). An important variation on the models shown in Figure 1a
and Figure 1b involves a neutralization process due to progressive water rock interaction from acid
zone (vapor core) toward productive outflow (Ellis and Mahon, 1977; Allis et al., 2000; Moeck,
2014).
Vapor Core Solfatara Fumarole
Vapor Core Neutral-Liquid Reservoir
Acid Cl-SO4
Solfatara
HCO3-Cl
Acid Cl-SO4
Crater Crater
Solfatara Solfatara HCO3 Water
Na-HCO3 Water
2000 Acid Cl-SO4 2000 Acid Cl-SO4
Acid Cl-SO4 Acid SO4 Na-Cl Water
1500 1500 Fumaroles
HCO3-Cl
1000 1000 HCO3
Na-HCO3
Na-Cl
500 500
Boiling zone
0 Basin Fill Sediment 0

-500 -500
Unboiled Outflow
Base of Conductor Outflow
-1000 -1000
Unboiled Outflow
Neutral Upflow
-1500 -1500
Degassing Magma Degassing Magma
CO2, H2S, HCl, SO2, HF CO2, H2S, HCl, SO2, HF

Heat Source Heat Source

A) Single ‘Vapor Core System’ B) Paired Liquid & ‘Vapor Core System’

Figure 1. Conceptual model cross-sections of (a) a vapor core volcanic system and (b) a paired vapor core
system and neutral geothermal system (modified from Reyes et al., 1993).
2.1 Fumaroles
In exploitable geothermal prospects hosting neutral-pH hydrothermal fluids, the non-condensable
gas detected in fumarole samples indicating benign reservoir conditions are usually below total
NCG of 2 wt%, CO2 content typically 90-95% of the total gas, sulfur species dominated by H2S
with no SO2 gas, and no HCl gas (Table 1). Fumaroles with these characteristics are commonly
found above reservoir boiling zones, with the pattern of gas geothermometers and ratios of gases
with varying solubility providing a rough guide to whether the fumarole is located over an upflow
or outflow (Figure 1b).
In contrast, fumaroles associated with vapor core (magmatic) systems, like those in Figure 1a,
release gases that are compositionally different from neutral (sometimes called ‘mature’ or
‘equilibrated’) hydrothermal systems. Total gas content in vapor core fumaroles may be greater
than 10 wt% with significant concentrations of SO2 and HCl. In Indonesia, sulfur-depositing
solfataras are associated with the crater areas of active magmatic conduits within likely vapor core
zones. In some cases, like Kawah Putih at Patuha (Layman and Sumarinda, 2003; Ashat et al.,
2019), acidic crater lakes are perched far above the regional water table by the very low
 Putra et al.

permeability shell that isolates the vapor core. Giggenbach (1987) describes the origin of elemental
sulfur that actively deposits inside these chimneys, where oxygen is absent, by the oxidation of
H2S by SO2 (rather than O2). Another key characteristic of the gas from fumaroles in vapor core
systems is that magmatic steam often has very low H2 concentrations because of reaction with
SO2.
Overall, fumaroles are more likely to have clear signs of juvenile magmatic character in their gas
composition near a volcano summit that has had a history of repeated eruptions, whereas the
magmatic character of the gas chemistry becomes lower when moving away from the summit
toward lower elevation flank fumaroles. In Figure 1b, the flank fumaroles at lower elevation have
a more benign chemistry consistent with steam and non-condensable gas produced from boiling
of a neutral.

2.2 Chloride hot springs

Neutral chloride hot springs are found where hot water buoyantly flows from a geothermal
reservoir through a permeable path that meets the water table at the surface. In the case of volcano-
hosted geothermal reservoirs, that is usually situated on the lower flanks of the volcano, offset
from the neutral geothermal reservoir upflow by 1 to 20 km, or more (Goff, 1988). The chloride
springs can be used to help constrain not only the reservoir temperature and chemistry but also the
depth of the top of the liquid reservoir (Cumming, 2016). The lack of chloride springs in a
geothermal system could imply that the water table is deep and so emerges at a large distance from
the resource, or possibly that the reservoir is isolated from the surrounding aquifers, as is always
the case for steam-dominated reservoirs like Kamojang and Darajat that leak gases and steam but
no chloride water (Grant and Bixley, 2011).
Table 1. The characteristics of convective neutral chloride systems versus vapor core (magmatic) systems.
Key Aspect of Convective Geothermal System
Vapor core (Young System)
Conceptual Model (Neutral-Fumarole)
Fumarole/Solfatara  Low NCG content (<2 Wt%)  High NCG content (>10 %) (with no
(Gas Geochemistry)  CO2 content 90-95 % without SO2 evidence of steam condensation)
gas  Significant concentration SO2
 No presence HCl or HF gas  HCl and HF gas presence from
 3He/4He value 5-7 Rc/Ra for magmatic vapor
volcanic arc system  High 3He/4He value (>7 Rc/Ra)
 Small shift δO18  δ2H and δ18O shift toward andesitic
 Higher methane concentration water
 Low methane concentration (<1%)
 High N2/Ar ratio (close to magmatic
N2/Ar ~800 and CO2/N2 ~200 line
from N2-CO2-Ar
Chloride Hot Springs  Boiling chloride springs >500 ppm  No boiling chloride springs, Cl-SO4,
(Liquid Geochemistry) Cl-HCO3 water – Cl from HCl
magmatic vapor
Clay Cap  Base of conductive zone below  Clay cap downdip following
(MT Resistivity) liquid level (Cl Springs) topography
 Buoyancy outflow  Advective flow (Sub-boiling)
 Low resistivity is associated with  Low resistivity is associated with
smectite clay alteration kaolinite alteration
 Putra et al.

Chloride hot springs can be associated with vapor core systems but, aside from their proximity to
a magmatic vent and related fumaroles, they also have chemistry that differs diagnostically from
the chemistry of chloride springs associated with neutral geothermal reservoirs. In a vapor core
system, the acid-chloride-sulfate (Cl-SO4) water is formed by the dissolution of volatile gases
(HCl, H2S and SO2) in steam condensate and meteoric water within the high gas, low pH
environment of the narrow, sometimes superheated, and vapor chimney above the magmatic
intrusion (Reyes et al., 1993).

2.3 Vapor Core System vs. Neutral Geothermal System

The differences between the conceptual elements of a neutral system and a vapor core geothermal
system are illustrated by the differences in the expression of surface manifestations that guide
exploration. In the upflow zone of the distributed-permeability geothermal reservoir shown in
Figure 1b, the high-temperature water flows almost vertically upward by buoyancy and, if the
upflow reaches the boiling point, gas and steam tend to leak through the clay cap, appearing at the
surface as acid-sulfate fumaroles. The base of the low resistivity smectite clay cap over such a
boiling zone is typically a bit higher than the local water table. The impermeable smectite clay cap
then directs water laterally into an outflow zone where buoyant flow continues up-dip beneath the
clay cap, emerging at a neutral chloride or chloride-bicarbonate hot spring. The top of the vapor
core system is commonly associated with fumaroles that may contain HCl or SO2 gases diagnostic
of the underlying vapor core unless these gases are dissolved in perched meteoric water along with
H2S. The resulting acid chloride sulfate water flows down the volcano flank on top of the low-
resistivity kaolinite and smectite clay alteration, typically emerging as hot springs with similar
chemistry but probably not fumaroles.

3. Vapor Core Conceptual Model of Mt. Talang Prospect

Following the description of geoscience data used to characterize the Mt. Talang prospect and the
ambiguous implications of the data set considered independently, the results are integrated to
assess how decisively they differentiate potential vapor core and neutral reservoir zones as part of
a conceptual model.

3.1. Geology and Tectonic Setting

The Mt. Talang prospect, outlined in Figure 2 , lies on the Great Sumatra Fault Zone (GSFZ)
40 km east of the city of Padang, West Sumatra Province. The GSFZ is highly segmented and fault
segments vary in length from about 35 to 200 km seperated by major fault discontinuities
(Natawidjaja, 2017). The working area outline includes the active volcano of Mt. Talang (2597 m)
in the south and the structural pull-apart basins of the GSFZ to the north at 400-450 mrsl (metres
relative sea level), which are bordered by basement rocks.
The Mt. Talang prospect is associated with two en echelon pull-apart basins, both associated with
a ~10 km right step of the Suliti Fault in the south to the Sumani Fault in the north (Figure 2).
Subsurface fault permeability associated with the fumaroles is consistent with left-lateral
kinematics that include a significant trans-tensional normal component in the fumarole structure
across Mt. Talang (Figure 2b). The basins are in-filled by volcanics and bordered by Mesozoic
meta-sediments and Paleozoic sediments. The original caldera rim is still visible on the south and
 Putra et al.

east side of the volcano. Young andesitic lava then deposited within the collapsed caldera and
produced numerous lava flows that reached up to 15 km to the NNE.

Group 8

Figure 5

Group 7
Solok
Basin

Group 6

Talang
Basin
Group 4
Legend Group 5
Fumarole Group 3
Group 2
Solfatara
Group 1
HCO3 Spring Mt. Talang
Cl-HCO3 Spring Dibawah
Cl Spring
Lake

Fault

Lineament

Collapse / rim

Mountain

(a) (b)
Figure 2. (a) Index maps show the location of Mt. Talang in Sumatra and the regional-scale fault map highlights
the main structural framework of Mt. Talang in the Great Sumatra Fault Zone (GSF) (Sieh et Al., 2000;
Natawidjaja, 2017). (b) Thermal feature map of the Mt. Talang geothermal prospect classified into eight
groups.
3.2 Geochemistry
3.2.1 Gas Chemistry (Fumaroles)
The solfatara (Group 1, surface temperature 92.8°C) located near the summit has a high N2/Ar
ratio. The presence of SO2 indicates magmatic vapor lies beneath the summit. Gas concentrations
in the fumarole steam from all localities range from 2.5 to 10 wt%, with some higher gas
concentrations potentially caused by steam condensation and dissolution of soluble gases. The
gases from the Mt. Talang fumaroles show a trend to the NE, from compelling indications of a
vapor core in Group 1, to having an ambiguous affinity to a vapor core in Group 3, to general
indications of an underlying benign geothermal reservoir in Group 4. Methane (CH4 in Figure 3a)
concentrations are low (<1%) in the summit Group 1 and Group 3 flank fumaroles but are 5-10
times higher in Group 4. It indicates the equilibration at lower temperatures in more benign
hydrothermal conditions when moving to the northeast from Group 1 (Giggenbach and Glover,
1992). The Group 1 and Group 3 fumaroles have near magmatic N2/Ar ratios (~ 800) (Figure 3b).
Based on gas trends from commercial geothermal systems (Nicholson, 1993), the Group 4
fumaroles that have higher CO2/N2 ratios, consistent with less air contamination, have a CO2-N2-
Ar distribution indicative of a high-temperature equilibrated liquid reservoir (Giggenbach, 1987).
The CO2-N2-Ar distribution for the Group 2 fumaroles suggest they are strongly influenced by
shallow air-saturated water (ASW) and so their interpretation with respect to deep processes will
 Putra et al.

be more ambiguous. The Group 4 fumaroles show a progression of increasing NH3 to the northeast
on the CO2-H2S-NH3 ternary (Figure 3c). This suggests the fumaroles are aligned along a zone of
boiling related to a neutral liquid system, where the fumaroles with the lower NH3 are closer to the
liquid upflow and the fumaroles with higher NH3 are located along a boiled outflow.

200 CH4 100 N 2

a) 160
180
200
b)
220
Magmatic
Gas Solubilities 90% (N2/Ar = 800)
240
N2 < O2 < H2 < CH4 < CO2 < H2S <
NH3 80%
260

70%

280 60% Near-magmatic

N2/Ar
50%
High Temp Deep Air (N2 /Ar = 84)
300
Upflow (liquid)
40%
Lower Temp
Shallow Steam
320 30% ASW (N2 /Ar = 38)

340 20%
360
10% Boiling – Loss
380
400 of volatile

CO2 10 H2S CO2 10000 Ar

Least Solube
c) 0.01 CO2
d)
5

90% 4
equilibrated vapor
80% GRID RH = -2.8
3
70%
equilibrated liquid
log(H2 / Ar)

60% 2

50%
1 Argon Error
Fumarole assoc. 40%
with Commercial 0
30%
Reservoir
System Gases
20% Reboiling -1
Outflow
Intermediate10%
-2
Solubility Most Solube 0 1 2 3 4 5 6 7 8
H2S 10 NH3 log(CO2 / Ar)

: Group 1 (Solfatara) : Group 2 (Fumarole) : Group 3 (Fumarole) : Group 4 (Fumarole) : Group 8 (Cl Springs)

Figure 3. (a) CO2-CH4-H2S ternary diagram; (b) CO2-N2-Ar ternary diagram; (c) H2S-CO2-NH3 ternary
diagram; (d) HAr-CAr gas grid.
aGas geothermometry from equilibrated fumaroles (red and green circle symbols that plot on the
gas grid in Figure 3d) support reservoir temperature estimates ranging from 250 °C to 310 °C. The
gas sample (red circle) from Group 4 that plots below the liquid equilibration line of the ratio cross-
plot H2/Ar – CO2/Ar may be impacted by hydrogen re-equilibration at lower temperature in its
path to the surface, argon gain from air or meteoric water, or an Rh value <-2.8 (Giggenbach and
Goguel, 1989). However, overall, the position of the Group 3 and Group 4 fumaroles on the liquid
equilibrium line in Figure 3d suggests that the fumarole steam formed from boiling of an
equilibrated reservoir liquid. The trend of decreasing gas content to the NE is important to the
interpretation of neutral liquid reservoir conditions at Mt. Talang. In vapor-dominated systems,
gas content increases towards the margins of the resource because of condensation of steam (Allis,
2000).
 Putra et al.

3.2.2 Liquid Chemistry (Hot Springs)

The anion distribution in Figure 4a shows four water types: bicarbonate-sulfate waters (Group 5),
bicarbonate water (Group 6), mixed bicarbonate-chloride waters (Group 7), and chloride waters
(Group 8). The four samples in Group 8 are more than 20 km to the NW of the Mt. Talang summit
and the hot springs have the highest chloride content of all the springs associated with Mt. Talang,
from 1264 to 1485 ppm chloride, which suggests that these springs are the best connected to a
chloride water source. There is no closer candidate volcano source for these springs other than Mt.
Talang, and the MT resistivity pattern, as well as the position of the Group 7 springs, indicates that
Mt. Talang is a plausible source of the chloride water, despite the distance and these springs being
50 m higher elevation than intervening mixed bicarbonate-chloride springs. The HCO3 (Figure 4a)
and Mg content of the Group 8 springs suggests they are mixed with at least 25% shallow
groundwater and/or on the margins of the geothermal system (Nicholson, 1993), and an evaluation
of liquid geothermometry (Na/K and K-Mg geothermometers) for springs along the long outflow path
suggests re-equilibration at lower temperatures (from 220 to 250 °C) for the Group 8 and Group 7
springs. The key significance of the Group 7 and Group 8 springs is that they suggest the presence
of a long outflow path from a benign neutral chloride system.

-40
Cl
-50
+
local meteoric
+
-60 water
Reservoir ?
-70 +
δ2H (‰)

primary steam
-80

-90
Secondary steam
-100 from steam-heated
groundwater
-110
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 -0
SO4 HCO3 δ18O (‰)
(a) (b)
Legend 10
R&S MORB
ARC
(RA)measured

: Group 1 (Solfatara)

: Group 2 (Fumarole)
1 AIR
: Group 3 (Fumarole)
3He/4He

: Group 4 (Fumarole) Sumatra-Gas

Sumatra-Water
: Group 5 (HCO3 – SO4 Spring) 0.1 IBVC-Gas
IBVC-Water CRUST
Java/Bali-Gas
: Group 6 (HCO3 Spring) Java/Bali-Water

: Group 7 (Cl-HCO3 Spring)

0.01
1 10 100 1000 10000
: Group 8 (Cl Spring)
X-Value
(c)

Figure 4. (a) SO4-Cl-HCO3 ternary diagram; (b) Stable isotope plot (c) Helium isotope ratios (Rm/Ra) versus
X values for the Mt. Talang compared to other locations in Sumatra, Java, and Bali. (Figure modified
from Halldórsson et al., 2012).
 Putra et al.

3.2.3 Stable Isotope (Oxygen - O18 vs Deuterium - H2) & Helium Isotopes
The deuterium δD and oxygen δ18O stable isotopic (Figure 4b) compositions from the Group 3
fumarole samples (δD = -60.5; δ18O = -5.8) follow a mixing trend between local meteoric water
and ‘andesitic water’ (water vapor from degassing magma) which has an inferred composition of
about δ18O = +10‰ and δ2H = -20‰ (Giggenbach, 1992), whereas the Group 4 fumaroles (δD =
-77; δ18O = -9.75) are consistent with primary steam derived from a liquid reservoir.
Figure 4c provides a comparison of Mt. Talang helium isotope ratios to values for other volcanic
centers in Sumatra, Java, and Bali using isotope data modified from Halldórsson et al. (2012). The
sample from the Group 1 solfatara at Mt. Talang has the second highest Rc/Ra (~7.76 Rc/Ra) value
for the Great Sumatra Fault region, suggesting high vertical crustal permeability and magmatic gas
input at this location, as would be expected at a vapor core system. Similar to the NCG samples,
the Group 4 helium Rc/Ra values vary widely (2.3 – 5.7) which may be due to varying levels of
mixing with air-saturated water at different fumarole sites.
3.2.4 Compartmentalization Between Neutral Chloride Reservoir and Vapor core
The geochemical data strongly shows two transitional trends between a magmatic and a neutral
liquid dominated system, which is interpreted as either a permeability barrier
(compartmentalization) of anhydrite/calcite mineralization in vapor core shell or neutralization
processes from an acid to neutral environment. The presence of a permeability barrier is supported
from methane and CO2-H2S gas ratios, where methane concentrations are low (<1%) in the
magmatic summit solfatara but are 5-10 times higher in northeast fumaroles, indicating
equilibration at lower temperatures in more benign hydrothermal conditions. The trend of
decreasing gas content to the NE is central to the interpretation of boiling outflow from a neutral
liquid reservoir.

3.3. Geophysics
The magnetotelluric method is applied in Mt Talang geothermal prospect. Hitay collaborated with
PT TBU and CGG (Milan) to acquire 84 stations (Figure 5) in 2017 and compute a 3D MT
inversion resistivity model. The CGG RLM-3D MT finite difference code (Soyer et al., 2017) was
used to invert the full tensor MT impedance, including topography, using a ~30 m digital elevation
model. One dimensional (1D) Occam inversions of the invariant and TE MT modes is used to test
the resolution and reliability of the 3D resistivity model (Cumming and Mackie, 2010), after
correcting the MT data for static distortion using topographic correlation or averaging, and
masking data distorted by noise or 3D effects inconsistent with the 1D inversion.
As expected, the 3D and 1D inversion cross-sections are mostly consistent in resolving a domed
low resistivity smectite layer with a base at a depth of <1000 m (Figure 7). The low resistivity
smectite cap is domed up to the highest elevation and is thinnest near the Group 3 and 4 fumaroles
(Figure 6). The thinner clay cap imaged by the 1D inversion, in comparison to the 3D inversion,
is characteristic of the performance of these methods in imaging the shallow clay cap of a
geothermal field near significant sources of electrical noise (Cumming, 2010). Fortunately, as
illustrated in Figure 6, the general shape of the base of the clay cap with respect to identifying an
upflow and likely path of buoyant outflow is similar in both cases. The connection of buoyant flow
from the upflow to the outflow connection to chloride springs is illustrated in Figure 8.
 Putra et al.

The geometry of the MT resistivity pattern in Figure 7 is consistent with a clay cap over a neutral
geothermal system, which indicates geothermal outflow buoyantly up-dip beneath the clay cap
(Cumming, 2016). Several aspects of the resistivity pattern are important to the discrimination of
the acid and neutral zones in the conceptual model. The rough alignment of the trend in the apex
of the base of the conductive clay cap with the NW-striking Fumarole Fault (Figure 6) suggests a
correlation of surface structure with underlying permeability. The zone in Figure 8, where the base
of the conductive clay cap is much higher above the regional water table near the solfatara at Mt.
Talang, is more likely to be associated with an underlying vapor core. The manner in which the
low resistivity zone follows topography down the flank of the volcano from the solfatara could
correspond to shallow downhill flow of acidic fluids rather than a clay cap over an underlying
reservoir.

Solok
Basin

Figure 6

Talang
Legend Basin
Fumarole D B
Solfatara

HCO3 Spring E
Cl-HCO3 Spring
A
Cl Spring
Mt. Talang Dibawah
Fault
Lake
Lineament

Collapse / rim

Mountain

MT Station 2011 & 2013

MT Station 2017

Figure 5. MT stations distribution in Mt. Talang. The red lines indicate the MT resistivity cross sections used
to illustrate the interpretation.
 Putra et al.

± ±
1D Inversion 3D Inversion

Group 6 Group 6

BOC 3-D
Group 4 (meter) Group 4
Group 5 Group 5

BOC 1-D
(meter)

Group 3 Group 2 Group 2

Group 3

Group 1 Group 1

Legend Legend
Fumarole Fumarole
Solfatara Solfatara
Hot Springs Hot Springs
m Fault m Fault
0 1.000 2.000 0 1.000 2.000
Doming MT Doming MT

Figure 6. The elevation of the base of the low resistivity zone 1D inversion (left) and 3D inversion (right) (in
mrsl), relative to an outline of a perimeter labeled as the ‘Doming MT’ area.
3.4 Mt. Talang Conceptual Model
To support the assessment of the geothermal resource at Mt. Talang, a conceptual model has been
developed, as illustrated by conceptual model elements drawn on resistivity cross-sections with
annotated fumarole and spring chemistry (Figures 5-8). The conceptual model elements include
the interpreted isotherms, water table, fluid phase, acid zone, neutral zone, upflow and outflow.
The primary indicators of the existence of a neutral geothermal system in the Mt. Talang
geothermal prospect are the fumaroles with gas chemistry indicative of a high temperature neutral
zone adjacent to a magmatic acid/vapor core, and hot springs with mixed chloride water chemistry
with cations consistent with the likely existence of a neutral reservoir. The geology provides the
overall context for the model and highlights overall risk issues, like the relatively shallow older
rocks below Mt. Talang that are likely to focus the permeable reservoir zone. The MT resistivity
resolution of the clay cap constrains the overall buoyant flow of the geothermal fluids and,
therefore, likely resource targets and capacity estimates. The fumarole gas geochemistry suggests
that a magmatic vapor core below the Group 1 fumaroles is bordered by a boiling neutral liquid
system below the Group 4 fumaroles. The existence of mixed chloride-bicarbonate hot springs
along a plausible outflow path with liquid geothermometer temperatures that overlap with gas
geothermometry temperatures is supportive of the existence of a neutral 250-310°C reservoir on
the lower NE flank of Mt. Talang. The MT resistivity pattern is consistent with the interpretation
of a vapor core near the Mt. Talang vent, a separate neutral geothermal upflow below the NE flank
of Mt. Talang, and a boiled deep chloride outflow that travels >10 km as the system outflows
below a low resistivity cap extending to the NNW, as illustrated in Figure 8.
In the conceptual model shown in Figure 8, a high-temperature magmatic vapor plume ascends
from a degassing magma body at depth below the Mt. Talang edifice along previous eruption
 Putra et al.

conduits. The gas discharged through the Group 1 solfatara is characterized by 10 wt% gas,
comprising mainly of CO2 but including SO2 and HCl that are consistent with a magmatic vapor
core. Heat loss at the margin and top of the ascending magmatic gas column causes steam
condensation at shallow depth and, with addition of meteoric water, gases dissolve in the cooled
fluid to form acidic water containing H2SO4 and HCl. This acid chloride-sulfate water flows down
in perched aquifers on a clay alteration zone and mixes with meteoric recharge in shallow aquifers,
forming acid chloride-sulfate waters that flow down topography on the flanks of Mt. Talang. This
low resistivity layer overlain by acid-sulfate-chloride springs has been often misinterpreted as a
clay cap covering a neutral geothermal reservoir in a single vapor core setting, thus the presence
of the thermal features that indicate the neutral system are critical to reducing exploration risk.
4 4
D E D E

3 MT Doming 3 MT Doming

2 Group 4 2 Group 4 Group 5

Group 5

1 1
BOC
Elevation (km.asl)

Liquid Level Liquid Level

Elevation (km.asl)

BOC
0 0

2500 C 2500 C

-1 -1

Legend Legend
-2 Fumarole -2 Fumarole
Hot Spring Hot Spring

(1D Inversion) MT Station (3D Inversion) MT Station

-3 -3
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
Kilometers Kilometers

Figure 7. The D-E cross section (location indicated in Figure 5) with optimistic isotherm (1D inversion) and
most likely isotherm (3D inversion) as preferred model.
The preferred model presented here for the acid and neutral zones on Mt. Talang geothermal
system is that the interpreted neutral chloride system on the NE flank of Mt. Talang is separated
from the vapor core by a permeability barrier consisting of clays derived from acid alteration and
mineral precipitation characteristic of mixing of acid sulfate-chloride and neutral chloride water
types, typically including anhydrite. The gases and vapor from the magmatic chimney will dissolve
and condense in the liquid zone marginal to the vapor core to create acidic fluid. The water-rock
interaction will enrich the liquid composition reflecting the acidized country rock. The high sulfate
and carbonate concentration in the condensate that initially down-flows in the zone adjacent to the
magmatic/vapor core will seal the margins with minerals like anhydrite due to their retrograde
solubility (Reyes et al., 1993 and Allis, 2000), isolating the vapor core from the adjacent neutral
system. However, an alternative model that might result in a much smaller reservoir would assume
that condensate from the magmatic vapor core is more gradually neutralized as the water interacts
with the country rock, providing a progressively more benign chemistry along an outflow path,
perhaps as indicated by the difference in gas chemistry between Group 3 and Group 4 fumaroles.
The implications of the single vapor core or a paired vapor core with a neutral system model will
have different impacts on the long-term development viability. The conceptual model suggesting
 Putra et al.

a permeability barrier between the vapor core and the neutral reservoir will have a reduced risk of
acid migration due to pressure reduction during production, although there is still some chance of
acid fluid incursion if the seal is compromised. There are commercial paired neutral and vapor
core fields, such as Patuha, that have been operating with no serious acid migration after the
reservoir has been produced for five years (Ashat et al., 2019).
Manifestation GROUP1 GROUP 3 GROUP 4 GROUP 5 GROUP 6 GROUP 7 GROUP 8
Estimated Reservoir 325°C 310 °C 275 - 250 °C 157 - 160 °C 161 °C
Temperature (°C) Boiling zone (steam cap)
Upflow (250>>> ⁰C)
4
A B
Ohm.m
Steam Cap

C
400
3 D-E Cross Section Tubular Outflow (140-180 ⁰C) 300
200
Discharge chloride Springs 150
2 100
Elevation (km.asl)

70
50
1 40
30
20
0 Liquid Level Sulfate Water 15
10
7
-1 5
Unboiled Outflow 4
3
-2 2

-3

-4

-5

-6
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
Kilometers

Vapour Chimney Solfatara HCO3 Spring Cl Spring Liquid Level Advention Flow
Symbols:
Steam Zone Fumarole HCO3 - Cl Spring Isotherm (0C) D-E Intersection Hydrothermal Fluid Flow

Figure 8: SE-NW cross-section A-B-C showing resistivity values from 1D model (optimistic) and interpreted
isotherms. Note the upflow location center of Group 3 fumarole which flows north to northeast and
discharges in Group 7 Cl springs.

5. Conclusions
The conceptual model presented for Mt. Talang has been useful in highlighting a common
exploration pitfall for vapor core systems - the interpretation of the low resistivity clay zone
blanketing the flank of a volcano as a potential cap of a neutral geothermal reservoir. Subtle
features of the clay cap, such as an up-doming base of the conductor and the presence of thick
alteration zones above the water table located in areas of fumarole activity, are key features for
distinguishing the different clay alteration geometries for paired vapor core-neutral liquid systems
versus a single vapor core system. The presence of acid sulfate-chloride springs alone is not useful
to distinguish a vapor core from a neutral system, as they may be present in both environments.
One conceptual uncertainty for the proposed paired vapor core and neutral liquid system is whether
the neutral system is separated from the vapor core by a low permeability shell, or if the neutral
system is formed by progressive neutralization of magmatic fluids that outflow from the vapor
core. The upside potential is more favorable for the low permeability shell model, which may
provide a barrier to incursion of acid fluids into the neutral reservoir with production drawdown.
Future work to better resolve the conceptual model in Mt. Talang includes exploration drilling to
confirm the reservoir temperature and neutral chemistry, correlate well alteration profiles with MT
data, determine if there is any indication of acid fluids influencing the reservoir, and confirm that
the permeability and production characteristics are sufficient to support commercial development.
 Putra et al.

Acknowledgment

We would like to thank the management of Hitay Energy Holdings for the permission to publish
this paper. Although numerous geoscientists have provided technical support for this project and
promoted knowledge of geothermal and vapor core systems in Indonesia, with respect to
geochemistry issues, these especially include Tom Powell, Brian Lovelock, Luigi Marini, and Julfi
Hadi.

REFERENCES

Abiyudo, R., Hadi, J., Cumming, W., and Marini, L. “Conceptual Model Assessment of a Vapor
core Geothermal System for Exploration; Mt. Bromo Case Study.” Proceedings, The 4th
Indonesia International Geothermal Convention & Exhibition 2016, Indonesia (2016).
Allis, R., Moore, J.N., McCulloch, J., Petty, S., and DeRocher, T. “Karaha-Telaga Bodas,
Indonesia: A Partially Vapor-Dominated Geothermal System.” GRC 2000, USA (2000).
Ashat, A., Itoi, R., Pratama, H.B. “Updating Conceptual Model of Ciwidey-Patuha Geothermal
Using Numerical Model.” IOP Conference Series Earth and Environmental Science 2019,
(2019).
Cumming, W. and Mackie, R. “Resistivity Imaging of Geothermal Resources Using 1D, 2D and
3D MT Inversion and TDEM Static Shift Correction Illustrated by a Glass Mountain Case
History.” Proceedings WGC 2010, Bali, Indonesia (2010).
Cumming, W. “Resource Conceptual Models of Volcano-Hosted Geothermal Reservoirs for
Exploration Well Targeting and Resource Capacity Assessment: Construction, Pitfalls and
New Challenges.” Proceedings, Geothermal Resources Council Annual Meeting 2016, USA
(2016).
Ellis, A.J. & Mahon, W.A.J. “Chemistry and Geothermal Systems.” Academic Press, New York,
N.Y. USA (1977).
Gherardi, F., Panichi, C., Yock, A.,Gerardo-Abaya, J. “Geochemistry of the Surface and Deep
Fluids of the Miravalles Volcano Geothermal System (Costa Rica).” Geothermics 31, (2002),
91-128.
Giggenbach, W.F. “Redox Processes Governing the Chemistry of Fumarolic Gas Discharges from
White Island, New Zealand.” Appl. Geochem.2, (1987), 143–161.
Giggenbach, W. F. “Isotopic Shifts in Waters from Geothermal and Volcanic Systems Along
Convergent Plate Boundaries and Their Origin.” Earth Planet. Sci. Lett. 113, (1992), 495-510.
Giggenbach, W.F. and Glover, R.B. “Tectonic Regime and Major Processes Governing the
Chemistry of Water and Gas Discharges from The Rotorua Geothermal Field, New Zealand.”
Geothermics, 21, (1992), 121-140.
Goff, F., Shevenell, L. and Gardner., J. N., “The hydrothermal outflow plume of Valles Caldera,
New Mexico, and a comparison with other outflow plumes.” Journal of Geophysical Research,
Vol 93, (1988).
 Putra et al.

Grant, M.A., and Bixley, P.F. “Geothermal Reservoir Engineering – Second Edition.” Academic
Press. Burlington, USA, (2011), 359 pp.
Halldórsson, S. A., Hilton, D. R., Troll, V. R., and Fischer, T. P. “Resolving volatile sources along
the western Sunda arc, Indonesia.” Journal Chemical Geology (2012).
Hochstein, M. and Sudarman, S. “Indonesian volcanic geothermal system.” Proc. WGC 2015,
Melbourne, Australia (2015).
Layman, E.B. and Sumarinda, S. “The Patuha Vapor-Dominated Resource West Java.” Workshop
on Geothermal Reservoir Engineering Stanford University 2003, California, USA (2003).
Licup, A.C., Sarmiento, Z.F., Omac, F.C., Maneja, F.C., Chandra, V.R., Esberto, M.B., Villareal,
M.J.Z., Baltasar, A.S.J., Mulyani, S., Sari, P.P., Jhonny. and Juandi, D. “Geothermal 3D
Subsurface Modeling – A Case Study from Sorik Marapi Field, Indonesia.” Proceedings the
5th Indonesia International Geothermal Convention & Exhibition 2017. Indonesia (2017).
Marini, L., Fung, A.Y. and Sanchez, E. “Use of Reaction Path Modeling to Identify the Processes
Governing the Generation of Neutral Na-Cl and Acidic Na-Cl-SO4 Deep Geothermal Liquids
at Miravalles Geothermal System, Costa Rica.” Journal of Volcanology and Geothermal
Research 128, (2003), 363-387.
Moeck, I.S. “Catalog of Geothermal Play Types Based on Geologic Controls”. Geothermics 37,
(2014), 867-882
Natawidjaja, D.H. “Updating Active Fault Maps and Sliprates Along the Sumatran Fault Zone,
Indonesia.” Global Colloquium on GeoSciences and Engineering. IOP Conf. Series: Earth and
Enviromental Science 118 (2018) 012001, (2017), 1-11.
Nicholson, K. “Geothermal fluids: Chemistry and exploration techniques”. Springer-Verlag,
(1993), 263pp.
Ramos-Candelaria, M., Sanchez, D.R., Salonga, N.D. “Magmatic contribution to Philippine
hydrothermal systems.” Proc. WGC 1995, Firenze, Italy (1995).
Reyes, A.G., Giggenbach, W.F., Saleras, J.R.M., Salonga, N.D., Vergara, M.C. “Petrology and
geochemistry of Alto Peak, a vapor cored hydrothermal system, Leyte province, Philippines.”
Geochermics 22, (1993).
Reyes, A.G. “Petrology of Philippine Geothermal Systems and The Application of Alteration
Mineralogy to Their Assessment.” Journal of Volcanology and Geothermal Research 43,
(1990).
Sarmiento, Z.F., Bjornsson, G., Licup, A.C., Esberto, M.B., Indra, T., Baltasar, A.S.J., Omac, F.C.,
Maneja, F.C., Omac, X.L., Villareal, M.J.Z., Chandra, V.R. “Update on the Exploration and
Development Drilling at the Sorik Marapi Geothermal Field, North Sumatra, Indonesia.”
Proceedings The 5th Indonesia International Geothermal Convention & Exhibition 2017,
Indonesia (2017).
Sieh, K. and Natawidjaja, D. “Neotectonics of the Sumatran Fault, Indonesia.” Journal of
Geophysical Research: Solid Earth (1978–2012), v. 105, no. B12, (2000), p. 28295-28326.
 Putra et al.

Soyer, W., Mackie, R., Hallinan, S., Pavesi, A., Nordquist, G., Suminar, A., Intani, R. and Nelson,
C. “Multi-physics imaging of the Darajat field.” GRC Annual Meeting, Transactions, 41,
(2017), 1724-1741.
Stelling, P., Shevenell, L., Hinz, N., Coolbaugh, N., Melosh, G., and Cumming, W. “Geothermal
systems in volcanic arcs: Volcanic characteristics and surface manifestations as indicators of
geothermal potential and favorability worldwide.” Journal of Volcanology and Geothermal
Research 324, (2016), 57–72.
Stimac, J., Ganefianto, N., Baroek, M., Sihotang, M., Ramadhan, I., Mussofan, W., Sidik, R.,
Alfiady, A., Dyaksa, D.A., Aziz, H., Putra, A.P., Martikno, R., Irsamukti, R., Santana, S.,
Matsuda, K., Hatanaka, H., Soeda, Y., Cariou, L., Egermann, P. “An overview of the Muara
Laboh geothermal system, Sumatra, Indonesia.” Geothermics 82, (2019), 150-167.
Ussher, Greg., Harvey, Colin., Johnstone, Roy., Anderson, Errol. “Understanding the Resistivities
Observed in Geothermal System.” World Geothermal Congress 2000, Tohoku, Japan (2000).

View publication stats