Jim Lawless SKM Module 02 Systems PDF
Jim Lawless SKM Module 02 Systems PDF
Jim Lawless SKM Module 02 Systems PDF
Hydrothermal
Systems: The Basic Model
Module 2
PAGE 1
Contents
1. Basics of Hydrothermal Systems and Their Geological Setting 4
2. Boiling in Hydrothermal Systems ...............................................14
3. The Physical Role of Gas.............................................................15
4. Formation of Secondary Fluids in Hydrothermal Systems.......16
5. Practical Exercise.........................................................................18
PAGE 3
1.
The hydrothermal systems we are discussing are all related to magmatic activity. In this section we
shall go over the basics of such hydrothermal systems (Figure 1).
The type of hydrothermal system that can support economic geothermal energy developments in
Indonesia is where magmatic intrusions are emplaced high enough in the crust that they induce the
convective circulation of groundwater. They may or may not be related to eruptive volcanic activity.
There are other types of hydrothermal system such as the tectonic, or deep basin brines, but these are
irrelevant for our purposes as they are too cool, too deep or too saline to allow the economic
generation of electricity at this time in Indonesia. There are a number of sub-types of volcanic-related
hydrothermal system, which will be discussed in detail, but for now let us look at the basics.
Condensation
zone
Secondary
fluids
water gas
steam + gas water
Boiling
zone
water gas
Convective
upflow
2 - 5km
Thermal activity
tflow
Lateral ou
present)
(not always
ge
ar
ch
Re
magmatic
volatiles
Intrusive
PAGE 4
The heat source is an intrusion or intrusions: there is good evidence that the most important and longlived systems are those fuelled by repeated small intrusions (i.e. dykes or stocks) rather than a single
large pluton. The depth of emplacement varies depending on the geology, but is usually in the range 2
to 5 km. The composition of the intrusive can be anything from granite to gabbro but is most
commonly dioritic. The composition of the intrusive varies systematically with the geological setting.
The depth of emplacement and other characteristics of the resulting hydrothermal system vary
systematically with the nature of the intrusive.
The host rocks may be of any type, but given that these systems form in volcanic areas, the most
probable host rocks are volcanic. It is not uncommon for an old sedimentary/metamorphic basement
to be present at depth, but it is generally of low permeability. It may still be productive, but will
inevitably be more difficult to produce from since permeable zones will be restricted and more
difficult to locate. The permeability contrast at the interface between basement and younger cover can
be important in controlling the hydrology of the hydrothermal system.
There may also be a younger sedimentary cover. If this is a texturally immature facies, proximal to
source and derived from volcanic rocks, its composition and physical properties may not be too
different from its volcanic parents. If, however, there are calcareous sediments present, or older
metamorphosed marbles, these can have a significant effect on the composition of hydrothermal fluid
passing through them. The results are detrimental to development.
As a general rule, therefore, the easiest systems to develop will be where there is a thick volcanic or
volcanogenic sedimentary sequence, without basement being too shallow or too many limestones or
marbles present.
The size of a typical hydrothermal system depends on the geology and topography. In general terms,
the upflow zone may be in the range of 1 to 3 km2. The outflow zones can be as long as 20 km,
though these will usually be preferentially channelled in certain directions rather than surrounding the
whole system radially (Figure 2).
The fluid at the depth of economic exploitation is predominantly meteoric water, though the input of a
certain percentage of magmatic water and associated volatiles can be very chemically important
(Figure 3). In certain cases seawater may be the source fluid instead, and this introduces
complications to the chemistry. However, in most systems, the typical fluid which circulates through
the bulk of the hydrothermal system is a dilute brine, perhaps one-tenth to one quarter of the salinity of
seawater. It has a near-neutral pH, and a significant content of dissolved gases. At depth, in the nearintrusive environment where there is a strong magmatic influence, HCl and SO2 are the most
important. In most cases, however, this is at greater depth than geothermal wells penetrate. At
shallower levels CO2 is the most important and H2S is the next most common. (see Giggenbach, 1992,
for the reasons why: Figure 4). Note that there are, however, very important possible variations in the
chemistry, which will be discussed in detail later.
PAGE 5
k:geo\lec\min\98min\mod1\fig10.ppt
Meteoric water
Andesitic vapours
Seawater
Hydrothermal
system
Mid-ocean ridge
hydrothermal system
Mantle wedge
The
Slab
Convecting mantle
k:geo\lec\min\99min\mod13\ ai27a.ppt
PAGE 6
CO2
Geothermal Vapours
c
du
Re
SO2
H 2S
Ne
ut r
0.6
alis
atio
0.8
n
tio
0.4
Volcanic Vapours
0.2
< 300C
> 300C
0.2
0.4
0.6
0.8
100HCl
Figure 4 Plot of composition for vapours from andesitic volcanoes (after Giggenbach 1992).
The highest temperatures within an intrusive are up to those of magmatic melts, over 1000C. At
these temperatures rocks are plastic and unable to sustain fractures, so there is little potential for fluid
to flow. It is only when the rocks cool sufficiently to be brittle that fluid can penetrate. In the
convective part of a hydrothermal system temperatures are up to the critical point of water (374C for
pure water, but considerably higher for brines), but for reasons relating to host rock permeability and
silica solubility, temperatures over about 330C are not commonly encountered in active convective
hydrothermal systems down to 3 km vertical depth. In general the zone which is most accessible for
economic energy extraction is between 240 and 300C.
At depth, close to the heat source, the fluid may be effectively isolated from the surface.
Consequently very high fluid pressures can build up, to reach lithostatic, or temporarily even greater.
This is deeper than most geothermal wells penetrate. Above this level, the fluid is in pressure
connection with surficial water (albeit remotely), and pressures are controlled by hydrostatic effects.
At this level and above, the fluid is not stationary (or else it would cool conductively): it moves
dynamically in response to pressure gradients, which are caused by temperature differences. Hot
water is less dense than cold water and so it is displaced upward by surrounding colder water. The
system is thus a large convective cell. There is a central upflow zone, and a corresponding downflow
or inflow zone of recharge fluid. If the pressure gradients and topography are suitable, there may be
long lateral outflow zones.
PAGE 7
Within the deeper part of the upflow zone, the fluid is single-phase. It has a pressure gradient
controlled by the physical properties of water, modified by its chemical constituents. Above this, there
may be a zone where the fluid has boiled, or flashed. The pressure gradient in this zone may be
controlled by the density of steam ( gas) (a "vapour-dominated" zone), or may be intermediate
between that of steam and water: a sort of froth, referred to as a "two-phase" zone.
There are various aspects of vapour-dominated zones that can lead to some confusion:
1) Water content. Even in relatively "dry" vapour-dominated zones, there is still much liquid water
present. The term "vapour-dominated" refers to the pressure profile, not necessarily to the
physical state of the fluid present. A typical figure for unexploited vapour-dominated zones seems
to be around 30 % by volume water present, which means in excess of 95 % by mass. Water
saturation can rise as high as twice this figure and still maintain a vapour-dominated pressure
profile (Figure 5).
2) Water is much less mobile than steam. A well tapping a vapour-dominated zone will therefore
tend to produce nearly dry steam, even if there is significant water saturation in the reservoir
(Figure 5). As the reservoir dries out the steam produced may gain heat from the rocks and
become mildly superheated.
3) Wells penetrating a liquid-dominated zone can still produce fluid with excess enthalpy, or even
dry steam, provided the permeability in the formation is low enough (Figure 5). The extra
enthalpy is provided by heat from the rock, ie by boiling. Thus production of relatively small
flows of dry steam does not mean a vapour-dominated reservoir has been tapped.
4) A steam zone must be dynamic, or it would collapse through conductive heat losses and
condensation. There must be a continual upflow, or in rare cases a lateral outflow, of steam.
Physical reservoir modelling of steam zones suggests that there is a slow, downwards percolation
of liquid condensate. Except for where steam is vigorously escaping from a steam zone, as
through fumaroles, it must be bounded by a zone of condensation.
5) Because steam has much lower energy content per volume than water (despite its greater
enthalpy), the energy stored within a steam zone is no greater than that in a liquid-dominated zone.
This explains how the formation of steam zones is thermodynamically possible. Some authorities
consider that all natural steam zones must have started out as liquid-filled and gradually drained:
certainly in several steam-dominated reservoirs there is evidence preserved in the reservoir rock
for earlier higher-temperature, liquid dominated conditions (e.g. at The Geysers, Moore and
Gunderson: 1995)
PAGE 8
Steam,
< 240
H=1035 to 2600
Boiling level
Water only
Steam only,
or mixture
240 Water
H=1035
240 Water
H=1035
Steam,
< 240
H=2600
Boiling
Steam,
< 240
H=2600
Boiling level
Steam only
Steam only
240 Steam
and immobile
water
Total H=1115
240 Steam
and immobile
water
Total H=1115
Super heating-zone
PAGE 9
The temperature gradient within the deeper part is conductive. Above this, within the convective
zone, the pressure controls it, since it is limited by the boiling point of water at various pressures. The
limiting condition for water to be liquid is called the boiling-point-for-depth (bpd) gradient (Figure 6
and Table 1). This is a column of water that is everywhere just at boiling point: any pressure release at
any point will cause it to boil. This is of more than theoretical interest: in many hydrothermal systems,
the temperature and pressure gradients in the upflow zone are very close to the bpd curve. The
temperature gradient (i.e. increase in temperature with unit depth, not absolute temperature) is
therefore high near the surface, and lesser at greater depth.
Temperature (C)
100
150
200
250
300
350
0
200
400
Depth (m)
600
800
10% NaCl
(1.7M)
1000
Pure water
1200
1400
4.4% CO2
(1.0M)
1600
k:geo\lec\min\98min\mod1\fig11.ppt
Figure 6 Hydrostatic boiling point versus depth relations of hydrothermal fluids, showing the
contrasting effects of salinity and gas content (after Henley 1985).
PAGE 10
Pressure
(Bars abs)
Depth
(m)
100
105
110
115
120
125
130
135
140
145
150
155
160
165
170
175
180
185
190
195
200
205
210
215
220
225
230
235
240
245
250
255
260
265
270
275
280
285
1.01
1.21
1.43
1.69
1.99
2.32
2.70
3.13
3.61
4.16
4.76
5.43
6.18
7.01
7.92
8.92
10.03
11.23
12.55
13.99
15.55
17.24
19.08
21.06
23.20
25.50
27.98
30.63
33.48
36.52
39.78
43.25
46.94
50.88
55.06
59.50
64.20
69.19
0
2
4
7
10
14
18
23
28
34
41
48
57
66
76
88
101
114
130
147
165
185
207
231
256
284
315
348
383
422
462
507
555
607
662
721
785
853
PAGE 11
Temperature
(C)
Pressure
(Bars abs)
Depth
(m)
290
295
300
305
74.46
80.04
85.93
92.14
926
1004
1088
1178
Temperature
Hy
dr
os
ta
BPD
tic
Liquid
Condensing level
Isothermal
Vapour
Static
Boiling level
Steam
Liquid
m
Te
Depth
D
BP
r
tu
ra
pe
re
su c
es stati
Pr Hydro
Fluid
movement
e
Conductive: various
profiles possible
Lit
ho
sta
tic
k:geo\lec\min\98min\mod1\fig7.ppt
Pressure
In contrast, the temperature gradient in a vapour-dominated zone is very small. The zone is close to
isothermal (Figure 7). Because of the properties of water, such zones are often around 235 - 240C.
The temperature and pressure gradient corresponding to a boiling-point-for-depth situation is
sometimes referred to as a "hot hydrostatic" gradient. We can also identify a "hot hydrodynamic"
gradient. This is where hot fluid is upflowing sufficiently rapidly to lose significant energy to friction.
PAGE 12
In this case the temperatures and pressures can exceed boiling-point-for-depth (which is usually the
limiting case): that is to say pressures and temperatures increase more rapidly with depth than is
possible with a bpd gradient. Thus more energy is stored close to the surface than would otherwise be
the case, and there is a greater potential for both boiling and hydrothermal eruptions.
Although hydrodynamic gradients are often invoked to explain various reservoir effects, they have
rarely been measured in practice since these shallow zones are usually cased off in wells. One notable
exception to this, where gradients as large as 140 % of hot hydrostatic were measured, was reported by
White et al. (1975)
The minimum depth to encounter economic temperatures is limited by the boiling-point for depth
gradient, so geothermal wells are usually cased to at least 400m, and often much greater: a casing
depth of 1000-1500m is not uncommon. The maximum drilling depth is limited by the cost. Few
geothermal wells are drilled to greater than 3000m vertical depth, or more than 1500m horizontal
deviation.
PAGE 13
2.
Boiling in any fluid can take place as a result of two processes: reduction of pressure, or addition of
heat. It is unusual for heat to be added to a hydrothermal system, except in the special case of dyke
intrusion, which can cause violent boiling. But in general, fluid boiling in hydrothermal systems
occurs in response to pressure drop. This can be a more or less gentle, steady-state process, or it can
be abrupt.
The gentle, steady-state type of boiling happens as upflowing fluids reach a zone where the confining
pressure is no longer sufficient to maintain them in the liquid form. In this case there will be a more or
less constant boiling level at some depth, overlain by a vapour-dominated zone, and probably the
emission of steam from fumaroles at the surface. Creation of a steam zone at depth requires the
presence of a zone of limited permeability and a dynamic flow of fluid to provide the necessary
pressure drop. The extreme case of gentle, continuous boiling in a hydrothermal system is evaporation
from a hot spring.
Boiling may be a cyclic process, as in a geyser. In this case temperatures will rise within an
unconfined fluid column until it is close to boiling-point-for-depth throughout. If boiling then occurs
at some point, some fluid will be emitted from the vent, and the reduction of pressure allows boiling to
extend progressively deeper until the immediate fluid supply is exhausted. The fluid then gradually
re-accumulates and temperatures start to build up again. Most geysers have an eruption cycle of
between a few hours and few days and in many the cycle is quite regular. There is however an
interesting example in New Zealand, the Inferno Crater at Waimangu, in which the water level and
temperature rise until overflow occurs over a six-week cycle. In this instance overflow occurs without
violent boiling.
Boiling can also be more abrupt in response to a sudden pressure drop, for example from rock faulting.
This could cause hydrothermal eruption (near surface) or hydrothermal brecciation (sub-surface).
Fluid will rush into the fracture created and in the process of boiling become super-saturated with
various solutes so that secondary minerals are deposited. In this way veins may be formed, with
mineralogical indicators of boiling, such as adularia, wairakite and platy calcite, and fluid inclusions
indicating fluctuating two-phase and single-phase conditions. Rhythmical banding indicates regularlyrepeated episodes of boiling. The mechanism that induces these conditions is thought to be
hydrothermal fracturing, often with associated brecciation, allowing a rapid drop in pressure. On a
long time scale this process may be quasi-cyclic.
PAGE 14
3.
A large gas content in the hydrothermal fluid at depth will promote the formation of a two-phase or
vapour-dominated zone (Figure 8). As a low-gas fluid flows up, it will not boil until it reaches a point
where the pressure reduces to the saturated vapour pressure of water at the appropriate temperature. In
contrast, a high-gas fluid will effervesce as soon as the pressure reduces to that of the saturated partial
pressure of the gas. Little steam may be lost along with gas initially, but it will mean that the
overlying fluid becomes two-phase and hence of low density. This reduces the confining pressures.
The flashing front can then move down until equilibrium is reached, this is why geothermal systems
with a high gas content in the primary fluid, such as Ohaaki in New Zealand, have a deep boiling
level. Mahon et al. (1980) give a good discussion of the effect.
Temperature (C)
150
250
200
2W
t%
0.2m
CO i
2 n
Depth (m)
500
0.5m
1000
0.5
m
300
Na
Cl
Na
Cl
CO
2 in
0.5
mN
aC
l
1500
Figure 8 Relationship between temperature and the depth at which boiling will commence for
solutions containing varying amounts of NaCl and CO2 (modified from Louks 1984).
PAGE 15
4.
d
Oxygenate
groundwate
r
Gro
H2 O
und
+C
O
wa
te r
H+ +
ta b
le
HCO
PAGE 16
Carbon dioxide forms a mildly acidic solution, down to about pH 3. Hydrogen sulphide does also, but
more significantly, hydrogen sulphide can oxidise to form an acid sulphate solution. Providing there
is sufficient oxygen to allow full oxidation of the sulphide, solutions in the range pH 1- 2 can result.
Oxidation may occur through contact with air, through contact with near-surface weathered rocks, and
by contact with dissolved oxygen in groundwater. Biogenic effects of oxidation by bacteria are
thought to be significant near-surface.
The near-surface secondary geothermal fluids are therefore usually acidic, with varying amounts of
sulphate and bicarbonate. They may be systematically zoned around the upflow zone, as hydrogen
sulphide is more soluble than carbon dioxide. Thus there tends to be a central zone of sulphate-rich
fluids and a peripheral zone of bicarbonate-rich fluids. Because these fluids are oxidised and acidic,
they tend to take up cations from the host rocks, including magnesium, manganese, aluminium and
ferric iron.
PAGE 17
5.
Practical Exercise
Water in the upflow zone of a certain hydrothermal system has a boiling- point-for depth
temperature gradient. At a depth of 1088 m, it is at a temperature of 300C. If it is not (quite)
boiling, what minimum pressure is it at (refer to Table 1)?
..................................... bars
If the water rises by 100 m, what is now the minimum pressure to prevent boiling ?
..................................... bars
..................................... bars
How much does the temperature have to change (fall) over the same interval to continue to
prevent the fluid from boiling?
..................................... C
Now consider some fluid at 256 m and 220C rising 100 m. How much does the pressure
change in this case ?
..................................... bars
How much does the temperature have to fall to prevent the fluid boiling?
..................................... C
G
Comparing your answers to D and F, what does this tell you about the
readiness of fluid to boil at different levels/temperatures ?
.............................................................................................................
PAGE 18
Now refer to Figure 6. If the fluid at 1008 m was a 10 % NaCl solution rather than pure
water, how much shallower could it be without boiling ?
....................................... m
I
Still looking at Figure 6: if the fluid contained 4.4 % CO2 in solution, how much deeper
would it have to be to prevent boiling?
...................................... m
(NB: you cannot give an exact answer to this from Figure 6, but estimate a minimum amount at least).
PAGE 19