日本地熱学会誌 J.Geotherm.

第 33 号 第 2 号(2011) Res.Soc.Japan
79～85 頁 短 報 Vol.33.No.2(2011)
P. 79～P. 85

Improvement of Understanding of Vapor-dominated Geothermal Resources

Mineyuki HANANO✻

(Presented 1 July 2010 at RE2010, Received 27 November 2010, Accepted 21 February 2011）

Abstract
Vapor-dominated geothermal resources were first exploited for commercial power generation, because of the
easiness for utilization, i.e. wells in vapor-dominated geothermal fields produce dry steam alone. Understanding of
the resource was improved first by accumulation of field experience, examples and data. Then, several different
conceptual models were developed based on interpretations of the data. These conceptual models were tested and
validated by numerical modeling, and then evolved into the current form. These processes led to the current
understanding of the vapor-dominated geothermal resources. Thus, the key factors which contributed for the
improvement are the increase of field data and experiences both in quantity and quality, and the advancement of
numerical techniques. Among many vapor-dominated geothermal fields, real natural state of a reservoir has only been
confirmed at Matsukawa, to date.

Keywords: geothermal, vapor-dominated, reservoir, natural state, modeling

1. Introduction 2. Vapor-dominated geothermal systems

Among several types of geothermal resources,
vapor-dominated geothermal resources were first exploited
for commercial purposes, because of the easiness for
utilization, i.e. wells in vapor-dominated geothermal fields
produce dry steam alone. This paper first reviews past
pathways of the improvement of understanding of the
vapor-dominated geothermal resources, then discusses key
clues for the improvement and finally summarizes
remaining questions.
There are two definitions on the vapor-dominated
geothermal resources from different viewpoints;
1) In some part of a geothermal reservoir, especially below
a cap rock, there is a “vapor-dominated zone” in which
steam is the pressure controlling phase.
2) Most of wells in a geothermal field produce dry steam
without hot-water.
The first one focuses on a natural state of the reservoir
and is related to a model of White et al. (1971), which will
be described later. This is a definition from a scientific
viewpoint. The second one focuses on discharge in an

✻
Japan Metals and Chemicals Co., Ltd., 1-17-25 Shinkawa, Chuo-ku, Tokyo 104-8257, Japan
ⒸThe Geothermal Research Society of Japan, 2011

― 79 ―
exploitation state. This is a definition from a practical a
viewpoint, and is widely accepted in field development.
In this paper, an entire system is called a
“vapor-dominated geothermal system” from an
earth-scientific viewpoint. A part of the system from which
dry steam can be produced for commercial utilization is
called a “vapor-dominated geothermal reservoir”. A
geothermal system which includes a vapor-dominated
geothermal reservoir that enables commercial energy
development is called a “vapor-dominated geothermal
resource”. Fig. 1 Understanding of a geothermal reservoir in
Larderello, Italy in the 1960s (Sestini, 1970).
3. Increase of field data
In early stages of the history of geothermal utilization, understandings among development engineers were still
increase in field experiences and data first played a very limited in and around the reservoir depths at that time, some
important role in the improvement of understanding of researchers had clear ideas on a total circulation in the
vapor-dominated geothermal resources. geothermal system in Larderello (e.g. Celati et al., 1973;
A systematic development of a vapor-dominated Hayakawa, 1968).
geothermal resource was started in Italy since around the Matsukawa, Japan is also one of the vapor-dominated
1900s (e.g. Cataldi et al., 1970; Dickson and Fanelli, 2004). geothermal fields. Understanding of its reservoir has been
By around the 1960s, understanding of the resource had improved historically by increased data and re-analyses (e.g.
been improved to the following level (e.g. Cataldi et al., Hanano and Matsuo, 1990). Historical changes of models of
1970); the Matsukawa geothermal system has been described by
1) Well drilling revealed that geothermal steam can be Ozeki et al. (2001).
produced from subsurface fractured zones.
2) Well drilling also revealed that the distribution of 4. Conceptual model
fractured zones are restricted and are not present Though there might have been several conceptual
everywhere. They must be related to faults and/or foldings. models, the first systematic one which describes entire
3) Well drilling also revealed the existence of a cap rock picture of a vapor-dominated geothermal system has been
above the production formation. It is impermeable and acts proposed by White et al. (1971) (Fig. 2). This model is
as a cover to the reservoir which limits inflow from a based on observations at The Geysers and Larderello, and
shallow ground water zone. has been thought to be the best to date (e.g. Grant and
These findings have been brought out by experiences Bixley, 2011). There are three factors which are
in well drilling. Thus increase in field experiences and filed characteristic of the model.
data improved understanding of the nature of the resource. 1) There is a vapor-dominated zone, in which steam is the
As a result, success rate of drilling improved stepwise. Also,
ideas for selecting drilling targets have been improved.
Fig. 1 is a conceptual model of a geothermal reservoir
in Larderello, Italy in the 1960s. At this stage, it has been
understood that the origin of geothermal steam should be
meteoric water. Based on the results of many geothermal
wells and production histories, the geothermal steam has
been found to be stored mostly as hot-water (liquid phase)
in pores and produced as steam (vapor phase) from
fractures within rock formations, because the total
Fig. 2 Conceptual model of vapor-dominated geothermal
volume of produced steam was much larger than the systems by White et al. (1971).
estimated pore space in the reservoir. Though the common

― 80 ―
pressure controlling phase. It is porous heat pipe, in which
steam ascends and condensate descends.
2) There is a low permeability barrier at least at the top and
sides of the reservoir. This limits inflow into and outflow
from the reservoir. Thus it allows the reservoir pressure
lower than hydrostatic.
3) Below the vapor-dominated zone, there is boiling brine.
Though most of the factors have been confirmed, the
brine has never been confirmed by drilling in any
Fig. 4 Conceptual model of vapor-dominated geothermal
vapor-dominated geothermal fields to date *. systems, modified from White et al. (1971) for
There is another model from a different view point. natural state modeling (Ingebritsen and Sorey,
D’Amore and Truesdell (1979) explained chemical 1988). “High chloride water” is the same as
distribution (non-condensable gas concentration) in a “Boiling brine” in Fig. 2.

shallow part of a vapor-dominated reservoir by introducing

effects of conductive heat loss through the cap rock (Fig. 3).
They noticed that water-vapor in geothermal steam may
condense if there is conductive heat loss through the cap
rock, but non-condensable gasses may not condense and
remain in the steam. Thus, gas concentration increases
towards the margin of the reservoir from its center along its
flow.
A summary of the most recent understanding of the
vapor-dominated geothermal systems can be found in Grant
and Bixley (2011). Fig. 5 Results of natural state modeling of vapor-dominated
systems (Ingebritsen and Sorey, 1988). In both cases,
5. Quantification vapor-dominated zone evolved below the cap rock
(kl) at the end of the runs.
Advancement of technologies of numerical modeling
played a very important role in the improvement of
understanding of vapor-dominated systems. Ingebritsen and
Sorey (1988) tested the conceptual model of White et al.
(1971) by numerical modeling of heat and mass flow in
porous media (Fig. 4). They succeeded in reproducing the
vapor-dominated condition below the cap rock (Fig. 5).
They found the characteristics of the vapor-dominated
systems as follows;

Fig. 6 Relationship between near-steady state

vapor-dominated zone pressures and the
depth from the upper pressure boundary to
the base of the caprock (Ingebritsen and
Sorey, 1988). Pressures are in bars.
Fig. 3 Conceptual model of a vapor-dominated
geothermal reservoir by D’Amore and 1) The conceptual model of White et al. (1971) is valid.
Truesdell (1979). 2) The low permeability barrier both at the top and sides of
* the reservoir is essential to form the vapor-dominated zone
Though many field engineers and scientists noticed it and discussed it
in conversations, it has not been described explicitly in any reports to (cf. Low k in Fig. 4 and kl in Fig. 5).
date, to the best of my knowledge.
3) Permeability of the cap rock and surrounding barriers

― 81 ―
should be around 5 x 10-17 m2. Lower or higher permeability
may not form the vapor dominated zone.
4) Maximum pressure at the top of the vapor-dominated
zone is controlled by the depth to the top of the
vapor-dominated zone. It is not controlled by the maximum
enthalpy of saturated steam (Fig. 6).
5) Minimum heat input necessary to maintain the
vapor-dominated condition is a function of the depth to the
top of the vapor-dominated zone. It is around 1.5 W/m2
when the depth to the top of the vapor-dominated zone is
about 300 m (Fig. 7).
Among them, the existence and value of the low
permeability barrier are the most essential;
1) If the order of permeability of the barrier is higher than
5 x 10-17 m2, the reservoir has much better hydraulic
contact with active in and out flow between the Fig. 8 Reservoir pressure profile in Kamojang in the
reservoir and outside. Thus heat is removed from the 1970s (O'Sullivan et al., 1990).
reservoir and will lead the system much cooler
(Hanano, 1992). source may not be necessary for the formation of a
2) If the order of permeability of the barrier is lower than vapor-dominated system compared with that of a
5 x 10-17 m2, the vapor-dominated zone will become liquid-dominated system.
much thicker, because fluid leakage from the top of the When the heat input is doubled, a much thicker
vapor-dominated zone to the shallow ground water vapor-dominated zone has been created. Thus, a
zone through the cap rock is too small. This may not vapor-dominated zone could have easily extended deeper
reproduce the natural state of real vapor-dominated and shrunk with increasing and decreasing conductive heat
reservoirs (Ingebritsen and Sorey, 1988; Hanano, flux due to repeated intrusion of magmatic heat sources
1992). (Hanano, 1992).
Note that above mentioned heat input (1.5 W/m2) is Note that thickness of the vapor-dominated condition
lower than 4 W/m2 which is a typical convective heat flux in the natural state is around 100 m to 200 m (Fig. 5). This
within liquid-dominated geothermal reservoirs (e.g. Hanano is much thinner than that confirmed in Kamojang in the
and Kajiwara, 1999). Thus, larger and/or stronger heat 1970s (Fig. 8). The problem on the natural state of the
a Kamojang reservoir will be discussed later.
As shown above, quantification of conceptual models
by numerical modeling gave various insights for better
understanding of the nature of vapor-dominated systems.
For the success of such numerical modeling studies,
preceding analytical studies also played very important roles
(e.g. Schubert and Straus, 1979; 1980; Straus and Schubert,
1981). For example, Fig. 9 is a result of analysis on
permeability required for the cap rock (Schubert and Straus,
1980). This result shows that a liquid phase above a vapor
phase is stable against any wave length of disturbance when
the permeability of the cap rock is below 4 x 10-17 m2.
Fig. 7 Minimum heat input rates necessary to Based on this study, Ingebritsen and Sorey (1988) employed
sustain a vapor-dominated zone at various
5 x 10-17 m2 as the permeability for the cap rock and
depths below an upper pressure boundary
surrounding barriers.
(Ingebritsen and Sorey, 1988).

― 82 ―
 Fig. 9 Dimensionless growth or decay rate versus
dimensionless wave number for
perturbations of a condensate layer overlying
the steam zone of a vapor-dominated
geothermal system (Schubert and Straus,
1980).
Fig. 10 Natural state of the Matsukawa geothermal
reservoir (Hanano and Matsuo, 1990). There
was a thin vapor-dominated zone below the cap
rock. Current production zone was filled with
liquid. This model has been validated by
numerical modeling (Hanano, 1992).
The effectiveness of quantification to improve the
understanding of a vapor-dominated geothermal reservoir
has been clearly demonstrated in Matsukawa. Hanano and
Matsuo (1990) numerically analyzed data collected in early
stages of the development in Matsukawa and reconstructed
the natural state of the reservoir (Fig. 10). Then Hanano
(1992) numerically validated its feasibility.
6. Natural state
Shallow well drilling and utilization of steam was
started in the 19th century in Larderello (e.g. Cataldi et al.,
1970; Dickson and Fanelli, 2004), and in the 1920s in The
Geysers (e.g. Barker, 2000). Also, a shallow exploration
well in Kamojang has been blowing out steam since 1928
(e.g. Zen and Radja, 1970) (Photo 1). Numerical modeling
― 83 ―
Photo 1 Blow-out of steam from one of the first
exploration wells in Kamojang, Indonesia
(photo courtesy of West JEC). This well
has been blowing out steam since 1928
(e.g. Zen and Radja, 1970).
studies showed that long-term steam production can
increase the size of a vapor-dominated zone substantially
even though its rate is very small (e.g. Pruess, 1985). Thus,
these long-term steam productions have increased the
thickness of vapor-dominated zones in these fields. Thus
“natural state” of the Kamojang reservoir observed in the
1970s (e.g. Grant, 1979) was not a real “natural state”. That
was “current state” in the 1970s instead (Fig. 8).
On the other hand, Matsukawa geothermal reservoir
was discovered in 1952. Then, a systematic geothermal
exploration was started in 1956, resulting in commencement
of commercial power generation in 1966 (e.g. Mori, 1970;
Hanano et al., 1993). Thus, real natural state of a
vapor-dominated reservoir has only been confirmed in
Matsukawa (Hanano and Matsuo, 1990; Hanano, 1992)
(Fig. 10). Note that the thickness of the vapor-dominated
condition in the natural state is only 200 m, which agrees
with the result of Ingebritsen and Sorey (1988) (Fig. 5).
7. Conclusion and suggestions for further research
Understanding of vapor-dominated geothermal
systems has been improved first by collection and
accumulation of field experience, data and examples. Then
conceptual models were developed based on interpretation
of the data, and then their feasibility was tested by
numerical modeling. A series of these efforts led to the
current understanding of vapor-dominated systems. Among
many vapor-dominated geothermal fields, real natural state
of a reservoir has only been confirmed at Matsukawa, to
date.
 However, there still remain some questions on the
vapor-dominated geothermal systems;
1) Brine has never been confirmed in any vapor-dominated
fields. Does it really exist? Is it still necessary or not from
both scientific and engineering viewpoints?
2) One more factor, i.e. cap rock and surrounding low
permeability barriers, is needed for the formation of the
vapor-dominated geothermal systems compared to those of
liquid-dominated geothermal systems. Thus probability of
occurrence of vapor-dominated geothermal systems is
believed to be lower than that of liquid-dominated
geothermal systems. However, necessary heat input to
sustain vapor-dominated geothermal systems is lower than
that of liquid-dominated geothermal systems. Thus, is the
probability of occurrence of vapor-dominated geothermal
systems really lower than that of liquid-dominated
geothermal systems or not? If it is really lower, then how
low is it?
3) A porous heat pipe sometimes exhibits instability, e.g.
cycling of direction of hydrothermal convection, especially
at high Rayleigh Number (e.g. McGuiness et al., 1993).
Thus, there is a possibility that this cycling had occurred in
existing vapor-dominated geothermal fields. If this is the
case, is it possible to confirm this phenomenon in a real
geothermal field by mineralogical study and/or other
methods?
Acknowledgments
This work is the revision and extension of the paper in
RE2010 International Conference in Yokohama with the
paper number O-Ge-3-5. This opportunity is gratefully
acknowledged. The author is also grateful to Japan Metals
and Chemicals Co., Ltd. for their permission to publish this
paper and to anonymous referees for their helpful
comments.
References
Barker, B. (2000) The Geysers: Past and Future. Geothermal
Resources Council Bulletin, 26, No.5, 163-171.
Cataldi, R. Ceron, P., Di Mario, P. and Leardini, T. (1970) Progress
report on geothermal development in Italy. Geothermics, Special
Issue 2, U.N. Symposium on the Development and Utilization of
Geothermal Resources, Pisa 1970, 2, Part I, 77-87.
Celati, R., Noto, P., Panichi, C., Squarci, P. and Taffi, L. (1973)
Interactions between the steam reservoir and surrounding
aquifers in the Larderello geothermal field. Geothermics, 2,
174-185.
― 84 ―
D'Amore, F. and Truesdell, A.H. (1979) Models for steam
chemistry at Larderello and The Geysers. Proceedings of the 5th
Workshop on Geothermal Reservoir Engineering, Stanford
University, 283-297.
Dickson, M.H. and Fanelli, M. (2004) What is geothermal energy?
International Geothermal Association,
http://www.geothermal-energy.org/
Grant, M.A. (1979) Mapping Kamojang reservoir. Geothermal
Resources Council Transactions, 3, 271-274.
Grant, M.A. and Bixley, P.F. (2011) Geothermal reservoir
engineering, 2nd edition, Elsevier, 378p.
Hanano, M. (1992) Simulation study of the Matsukawa geothermal
reservoir: natural state and its response to exploitation. Journal of
Energy Resources Technology, 114, 309-314.
Hanano, M. and Matsuo, G. (1990) Initial state of the Matsukawa
geothermal reservoir: Reconstruction of a reservoir pressure
profile and its implications. Geothermics, 19, 541-560.
Hanano, M. and Kajiwara, T (1999) Permeability associated with
natural convection in the Kakkonda geothermal reservoir.
Geothermal Resources Council Transactions, 23, 351-360.
Hanano, M., Kotanaka, K. and Ohyama, T. (1993) A quarter
century of geothermal power production at Matsukawa, Japan.
Geothermal Resources Council Bulletin, 22, No.2, 32-47.
Hayakawa, M. (1968) Hydrothermal system at Larderello in Italy
and anomalous heat flow in Japan. Chinetsu, 15, 69-86.*
Ingebritsen, S.E. and Sorey, M.L. (1988) Vapor-dominated zones
within hydrothermal systems: evolution and natural state.
Journal of Geophysical Research, 93, 13635-13655.
McGuinness, M.J., Blakeley, M., Pruess, K. and O'Sullivan, M.J.
(1993) Geothermal heat pipe stability: solution selection by
upstreaming and boundary conditions, Transport in Porous
Media, 11, 71-100.
Mori, Y. (1970) Exploitation of the Matsukawa geothermal area.
Geothermics, Special Issue 2, U.N. Symposium on the
Development and Utilization of Geothermal Resources, Pisa
1970, 2, Part 2, 1150-1156.
O'Sullivan, M.J., Barnett, B.G. and Razali, M.Y. (1990) Numerical
simulation of the Kamojang geothermal field, Indonesia.
Geothermal Resources Council Transactions, 14, 1317-1324.
Ozeki, H., Fukuda, D. and Okumura, T. (2001) Historical changes
of geothermal reservoir models at Matsukawa, Japan. Chinetsu,
38, 339-368.*
Pruess, K. (1985) A quantitative model of vapor dominated
geothermal reservoirs as heat pipes in fractured porous rock.
Geothermal Resources Council Transactions, 9, Part 2, 353-362.
Schubert, G. and Straus, J.M. (1979) Steam-water counterflow in
porous media. Journal of Geophysical Research, 84, 1621-1628.
Schubert, G. and Straus, J.M. (1980) Gravitational stability of water 短 報
over steam in vapor-dominated geothermal systems. Journal of
Geophysical Research, 85, 6505-6512. 蒸気卓越型地熱資源の理解進展過程
Sestini, G. (1970) Superheating of geothermal steam. Geothermics,
Special Issue 2, U.N. Symposium on the Development and 花野峰行
Utilization of Geothermal Resources, Pisa 1970, 2, Part I,
622-648. （2010 年 7 月 1 日 RE2010 にて発表，
2010 年 11 月 27 日受付，
Straus, J.M. and Schubert, G. (1981) One-dimensional model of 2011 年 2 月 21 日受理）
vapor-dominated geothermal systems. Journal of Geophysical
Research, 86, 9433-9438. 概 要
White, D. E., Muffler, L.J.P. and Truesdell, A.H. (1971) 蒸気卓越型地熱資源は，蒸気のみを生産することから利用
Vapor-dominated hydrothermal systems compared with が容易であり，最も早くから商業利用が行われてきた。開発
hot-water systems. Economic Geology, 66, 75-97. の進展によるデータや経験の蓄積とともにその理解は進ん
Zen, M.T. and Radja, V.T. (1970) Result of the preliminary で行き，その成果として種々の概念モデルが逐次考えられた。
geological investigation of natural steam fields in Indonesia. その後，数値解析技術の発達に伴い概念モデルの検証が行わ
Geothermics, Special Issue 2, U.N. Symposium on the れ現在の形に至っている。現在の蒸気卓越型地熱資源に関す
Development and Utilization of Geothermal Resources, Pisa る理解はこのような過程を経た結果である。従って，現在の
1970, 2, Part I, 130-135. 理解に至るまでに重要な役割を果たしたのは，データや経験
* in Japanese with English abstract の質・量両面での充実と数値解析技術の発達であった。多く
の蒸気卓越型地域が存在するが，現在までのところ，開発前
の自然状態が明らかにされたのは岩手県の松川地域だけで
ある。

キーワード：地熱，蒸気卓越型，貯留層，自然状態，モデリ
ング

― 85 ―