M03 Permeability Controls

Permeability Controls inHydrothermal Systems

Module 3

Permeability Controls in Hydrothermal Systems Contents i

Contents

1. Why Permeability is Important 1

2. Sources of Permeability 2

2.1 Primary Permeability 2

2.2 Secondary Permeability 2

2.3 Relative Importance of Primary and Secondary Permeability to Fluid Flow and Mineralisation 6

2.4 Locations and Orientation of Secondary Permeable Zones 7

2.5 Variations in Permeability with Depth 10

2.6 Applications of These Principles in Exploration 14

/tt/file_convert/55cf903b550346703ba41ac3/document.doc

Permeability Controls in Hydrothermal Systems 1

1. Why Permeability is Important

A hydrothermal mineral deposit is a fossil record of fluid-rock interaction. For this to occur, fluid must pass through the rock. Unless the rock is permeable, little fluid can pass through and so there can be little mass transfer. Other things being equal, the more fluid that passes through, the larger and richer can be the mineral deposit formed. Any process of brecciation also exposes more rock surfaces to fluid-rock interaction, and thus speeds up the necessary chemical processes. The objective of this section is to identify which are the most important types of permeability for hydrothermal ore deposition, and hence see how an understanding of the geometric relations of permeable zones might be used to predict the localisation of mineralisation.

Porosity: distribution usually random

Cooling joints and autobrecciationin lava flows: either random or insub-horizontal zones

Lithological contacts:either sub-horizontal or at low to moderateangles of dip in draped pyroclastics

AI37/14.1M

Diatremes, volcanic vents: sub-vertical

Figure 1: Primary permeability in volcanic rocks



2. Sources of Permeability

2.1 Primary Permeability

Massive volcanic rocks are the most common host rocks for epithermal and porphyry deposits. The sources of primary permeability in these rocks are (Figure 1):

Primary pores: These have very little effect on the permeability of the rock as a whole unless they are well connected. In most rocks they add little to the permeability. We must distinguish between porosity and permeability. Porosity is important for fluid storage, but unless it is connected it does not contribute to bulk permeability. There are exceptional rocks where it does matter, e.g. pumice.

Cooling joints: In lava flows these will form zones of permeability, e.g. Breccias at the top and bottom of lava flows. These zones of permeability will originally be sub-horizontal. In plutonic or hypabyssal rocks, they will form zones of permeability, e.g. around the margins. They are just as likely to be vertical as horizontal (or anything in between).

Lithological contacts: These zones of permeability will generally originally be sub-horizontal, but some air-fall pyroclastics can be draped over pre-existing terrain at quite steep angles of dip: 20° is not uncommon. In plutonic or hypabyssal rocks, the contacts are just as likely to be vertical as horizontal (or anything in between.

Volcanic vent structures: These are a special case of autobrecciation. They are generally sub-vertical.

In volcaniclastics and sediments, the degree of permeability will depend on the grain size and packing. Tuffs, for example, may be quite permeable or quite impermeable. Such rocks may also have a large surface area per volume, and hence be prone to alteration. Large vertical eruptive vents (e.g. diatremes), are a special case of highly permeable primary channels in volcanic terrains. A good example of where this has permitted selective mineralisation is the Montana Tunnels deposit in the USA (Sillitoe et al. 1985).

2.2 Secondary Permeability

In the typical island arc setting, the main sources of secondary permeability are (Figure 2):



Fracturing due to dyke injection:slightly more likely to be insub-vertical zones

Thermal cracking: distribution random

Rock dissolution: distribution random

Faulting: more likely to be sub-vertical

Hydrothermal brecciation:more likely to be sub-vertical

AI36/14.2M

Figure 2: Secondary permeability in volcanic rocks

Dyke Injection: Permeability due to the mechanical emplacement of a pluton before the onset of a hydrothermal system can be considered along with primary permeability as above. Obviously, repeated intrusion of dykes during the lifetime of a hydrothermal system will also cause secondary permeability due to purely mechanical effects. But as will be discussed elsewhere, it will have far more important effects because of its disturbance of the fluid temperatures and hence pressures. Such events, while important, can be considered infrequent in comparison to other sources of secondary permeability.



Thermal Cracking: A factor which is important in the near-pluton porphyry environment, but not so much in the epithermal environment, is fracturing of rocks by thermal stresses. This may be fracturing of host rocks as they are heated by an adjacent pluton, or, more significantly, fracturing of the intrusive rock as cool groundwater encroaches into them. A surprisingly small thermal change can cause rock to fracture: as little as 75°C under favourable circumstances of stress. Enhancement of permeability in intrusives by thermal fracturing is therefore a positive feedback mechanism encouraging fluid circulation and cooling of the pluton. The more cooling, the more fracturing, and the more fracturing, the more fluid can circulate.

Rock Dissolution: Except in special cases, such as limestones, this is a relatively minor source of permeability in volcanic terrains. It is however very important for the formation of oil reservoirs. In contrast, permeability reduction by deposition of secondary minerals is a very important process in controlling the hydrology of hydrothermal systems.

Tectonism: Faulting and related fracturing. Both normal and reverse faults can produce large fracture zones, since lateral stresses are often high. In plate-margin settings, repeated movement on faults is the norm. Once a fault is created, it is easier for it to move again that for a new fracture to form in response to accumulating stress. Slightly misdirected stress will therefore enhance permeability as open space is created in order to accommodate this. Rejuvenated structures may therefore be particularly permeable. Conversely permeability can be lowered in some situations and consistently high permeability cannot be assumed to exist all along a fault.

"Hydrothermal" Brecciation: This is a particular type of dilatant brecciation, which will take place when the fluid pressure in a rock exceeds the sum of the minimum principal stress, plus the tensile strength. But in a fractured rock (which is often the case in an area of active tectonism), the tensile strength is often almost zero, and in the arc environment the minimum stress will often be horizontal. Hence fractures can open at relatively low pressures, nowhere near "lithostatic" (where that is defined as equivalent to the vertical rock load). Once open, a fracture tends to extend. This process is called "hydrofracturing". It can result in a "jigsaw" or "crackle" breccia.

The formation of such a fracture allows fluid to flow much faster than before, since it short-circuits the hydrological gradient of the fluid. It may reach the point where clasts are transported along the fracture, producing rounding and the formation of rock flour. If the pressure difference between the source and the sink becomes great enough, the fluid may locally change phase, i.e. "boil", "effervesce" or "flash". This process involves a volume change, and thus very vigorous and rapid fluid movement. It may therefore cause considerable breakdown of the clasts within a fracture, creating a large permeable conduit (Figure 3). Flow will continue until the fracture becomes blocked, or the fluid source is exhausted.



Tectonism and hydrothermal brecciation go hand in glove: one may initiate the other; the same zones tend to remain as foci for repeated episodes of brecciation of both types. It is often not clear whether any particular event is strictly tectonic or strictly hydrothermal. To some extent the distinction is artificial. Tectonic forces create a state of stress, within which fracturing may be induced as increasing fluid pressures reduce the effective confining stress, and so movement results.

Figure 3:Idealised development of hydrothermal vein and breccias


Milled, fluidised breccia:rounded clasts, much fineclastic matrix

Stockwork

Polymict breccia

Rounded clasts,matrix becoming abundant

(Late stage veins)

Angular clasts, little matrix

“Jigsaw” or “Crackle” breccia

Root zone

Incr

easi

ng fl

uid

flow

rat

e

K:geo\lec\min\99min\mod15\fig2.ppt

Dec

reas

ing

fluid

flow

rate

and

clas

t tra

nspo

rt


2.3 Relative Importance of Primary and Secondary Permeability to Fluid Flow and Mineralisation

Secondary permeability is generally much more important than primary permeability in forming epigenetic deposits. Occasionally it is the other way around: for example, consider the high-sulphidation mineralised tuffs at Nansatsu (Figure 4), and possibly some of the low-sulphidation mineralisation at Toka Tindung (Wake et al. 1996). But in most cases secondary permeability is more important. There are several reasons for this:

Figure 4:Schematic model, Nansatsu type hot-spring gold deposit

(after Bonham 1989)

Fluid flow is predominantly within major fractures of secondary origin. The fractures created by secondary processes tend to be larger, and more continuous, than those created by primary processes. The ability of a fracture to transmit fluid varies as about the fourth power of the width of the fracture (depending on the shape). Thus a few big fractures are much more important than many little ones.

With time, hydrothermal mineral deposition blocks primary permeable channels. Secondary channels keep being rejuvenated, whereas primary ones do not.


andesite flows

porous acid leached zone

kaolinite smectite zone

quartzalunitezone

hydrothermalexplosion

breccia

powdery acid leached zone

silicified milled breccia

pyroclasticrocks

feeder vein

hydrofractured vein stockwork

quartz kaolinite alunite zone

electrum, pyrite, enargite,luzonite, bornite

K:geo\lec\min\99min\mod15\fig1

0 m

100

200

300

400

andesite flows


Economic mineral deposits require focussing of the locus of deposition, to concentrate the minerals to economic grades. This especially applies to minerals of high value, such as gold. Thus the fact that secondary mineral channels tend to be the locus of repeated episodes of brecciation means that economic (as opposed to "background") mineralisation is concentrated in them.

The processes of secondary brecciation take place abruptly. They therefore can introduce rapid changes to the fluid pressures, and this is the best means of inducing gold deposition, as we have discussed.

Secondary permeability-creating processes are more likely to produce permeable channels with a high angle of dip, whereas many primary zones of permeability are comparatively flat-lying (i.e. effectively stratigraphically controlled) (Figure 2). Low-angle fractures are not so common in this setting, except perhaps in response to removal of lithostatic load during the erosional unroofing of plutons. Near-vertical channels are more effective in connecting zones of different hydraulic pressures, and hence inducing fluid flow.

2.4 Locations and Orientation of Secondary Permeable Zones

Primary zones of permeability, as mentioned, can have any orientation, but are often near-horizontal. Secondary permeability channels tend to be more nearly vertical. This is because of the relative stresses in a typical volcanic belt.

In the back-arc situation, stresses will be extensional, and simple gravity faults at angles in the range 60-75° dip will result, depending on the rock properties. More often in plate-margin volcanic belts, forces will have a strong lateral component, and are often compressional. This means that both the least and greatest principal stresses will be horizontal. Thus hydrofracturing will result in near-vertical normal fractures. The fractures created become the pathways for the hydrothermal fluids, thus forming the typical near-vertical hydrothermal vein deposits (Figure 5b and c).

The situation with both horizontal stresses exceeding vertical (Figure 5a), which is common at depth in metamorphic zones, is not common at comparatively shallow levels at tectonic plate boundaries, which is where epithermal and porphyry deposits occur. Thus low-angle fractures are not so common in this setting, except perhaps in response to removal of lithostatic load during the erosional unroofing of plutons.

Rupture in response to tectonic stress may be expressed as a series of conjugate shears with a high angle of dip, or often an en-echelon structure of tension gashes rather than a single planar fault. This applies over a wide range of scales. The concept of the extensional or dilational jog has been successfully applied to the localisation of epithermal deposits. The space created provides a locus for pressure reduction, and so flashing of the fluid and mineralisation can occur (Figure 6). Recognition of this structural situation also explains why some vein deposits appear to be "cut-off" by faults,



but the "missing" extension of the veins cannot be located (Figure 7). It is important to distinguish syn-mineralisation faulting from post-mineralisation faulting.

At a deeper level, the intrusion of magma to form porphyry deposits may be localised within "pull-apart" structures.

Figure 5:Initial stress distribution causing faulting: 1 = maximum, 2 = mean, 1 = minimum (compressive) stress (after Hills 1964)



Figure 6: Rupture arrest at dilational fault jogsRupture perturbation and arrest (from Sibson 1985, 1987)


k:geo\lec\min\99min\mod15\fig6.ppt

A: Pre-rupture; uniform fluid pressures

Arrest

Perturbation

B: Post-rupture; fluid influx into jogs


Figure 7:Interpretation of the Martha Mine, Waihi, as a strike-slip

dilational fault jog (after Wellman 1954, Sibson 1987)

2.5 Variations in Permeability with Depth

The vertical zone at which economic epithermal mineralisation may occur is controlled to a large extent by the chemistry of the fluids, and the nature of the boiling-point for depth curve for water, while porphyry deposits are localised near to their intrusive parent by temperature constraints. But there are some general mechanical considerations that also play a role, and mean that some levels in a hydrothermal system are more likely to have extensive permeable zones than others. Let us look at the mechanics of a typical hydrothermal system from the bottom up (Figure 8).


80°

70°

EASTFAULT

WESTFAULT

Mine Plan, #9 Level, 178m below sea level

?

0 300

70°

80°

70°

?

k:geo\lec\min\99min\mod15\fig7

N


Near surface zone:

permeability reduced by silica sealing and argillic alteration. May be locally enhanced by hydrothermal eruptions

permeability enhanced by tectonic and hydrothermal fracturing, minor reduction by mineral deposition.

200°C

330°C

600°C

Main productive reservoir:

Convective / conductive transition zone:permeability reduced by lack of fracturingabove brittle/ ductile transition, near-criticalmineral deposition

Near pluton environment:permeability temporarily enhancedby “second boiling”


Note: vertical scale variable, highly dependent on geological setting

Figure 8:Vertical permeability distribution in hydrothermal systems

Much of the information on permeability comes from drilling for geothermal energy in active hydrothermal systems. As the technology improved, geothermal wells were drilled deeper and deeper, into hotter and hotter conditions. Thus the energy output of the wells became greater. But this process led to diminishing returns. Below a depth of about 2500 m in typical active hydrothermal systems in andesitic terrain, experience in geothermal drilling has shown that permeabilities are usually not consistently as high as at shallower levels. This was disappointing to those interested in energy extraction.

Experience has shown that the best wells for energy extraction often penetrate zones in the temperature range 260 - 300°C, rather than higher temperatures. The answer to this puzzle is to be found in the nature of rock deformation at these temperatures. Above about 320°C, typical host rocks will deform sufficiently rapidly that fractures either do not occur, or become healed within a short time in comparison to the rate of stress accumulation due to tectonic plate movement. Thus major permeability is short-lived.

The same phenomenon is well known to structural geologists working in metamorphic terrains, who refer to this as the "quartz-feldspar brittle-ductile transition", and set the temperature limit at about the same level (320-340°C). But in metamorphic zones, the much smaller geothermal gradient means that this transition occurs much deeper: many



kilometres, where pressures are also much greater. In hydrothermal systems the transition zone is much closer to the surface. This is also reflected in the depth limit of small earthquakes: it is well known that the number of micro-earthquakes drops off below the transition zone (10-15 km in continental crust). Active hydrothermal systems are often quite aseismic compared to the surrounding areas.

Another factor causing a decrease in permeability at depth is silica deposition. Late-stage magmatic and hydrothermal fluids are usually close to saturated with silica, simply because they are in equilibrium with silica-containing rocks. But the solubility of silica depends both on the temperatures and the fluid state. Supercritical fluids can contain much more silica in solution than sub-critical water (Figure 9). Thus if fluid cools from near-magmatic supercritical temperatures down to subcritical, it becomes silica supersaturated and quartz is deposited. We cannot put a single figure on the temperature of this process, since the critical point for water solution is so dependent on the concentration of dissolved solutes. But at some level around a cooling pluton, there will be a zone where silica tends to deposit and reduce permeability. This probably helps in keeping separate the near-pluton magmatic fluid and the overlying freely-convecting fluid which is largely of meteoric origin.

Figure 9:Calculated solubilities of quartz in water up to 900°C at the indicated pressures (after Fournier 1985)


K:geo\lecture\min\99mins\mod15\fig3.ppt

0

1000

2000

3000

4000

5000

300

200

100 500 700 900

Si0

2 (

mg

/ kg

)

Temperature (°C)

30040

0

500

600

700

900

80

01000

bar

s


Permeability within the porphyry environment and the deeper part of the epithermal environment is therefore short-lived. Permeability can be created by faulting, by intrusions, by hydrofracturing, or thermal cracking, but it does not remain open indefinitely.

These effects are accentuated by the properties of water at high temperatures. At high temperatures water has very low viscosity. In that respect it more resembles a gas than the fluid we are familiar with, and so can penetrate even tiny fractures readily. But it also has very low density, so a large volume of fluid moving through the rock is in effect providing a relatively small mass for chemical reaction and mineralisation. This is not true for the hyper-saline brines formed in the near-intrusive environment by phase separation on depressurisation. They are very dense and may in fact be too dense to convect, thus concentrating porphyry mineralisation near to the intrusive.

Moving up the hydrothermal system, the zone from 300 to 240°C, corresponding to perhaps 1 to 3 kilometres depth, is often found to be very permeable. The rocks are strong enough that fractures can be formed, cool enough that they do not rapidly heal, and covers an interval over which there is not a drastic drop in solubility for most volumetrically significant common hydrothermal minerals. However, because of the fluid properties, this is below the level at which most epithermal mineralisation occurs. Although hydrofracturing is common, the fluid is under too great a pressure for major flashing into fractures to be common. There are however some notable exceptions to this, in special cases where some local peculiarity has permitted flashing to extend to a greater depth.

The zone above 1000 m depth, that is to say from a level of 240-260°C up to ambient conditions at the surface, is where vigorous hydrothermal brecciation, in some cases leading to surficial hydrothermal eruptions, is most probable. It is thus the main zone of high-grade epithermal deposition. The reason for this is easily seen from the boiling-point for depth curve: the relative pressure gradient is much greater near to the surface (Figure 10). It is also within this zone that gas can separate from the typical hydrothermal fluid and accumulate to set up the conditions for triggering hydrothermal eruptions.

This is also the level at which a hydrothermal system in high elevation terrain may become stratified, with layers of different chemistry developing. In turn this can lead to permeability reductions due to secondary mineral deposition. So the overall effect is probably to focus the most important fluid flow in relatively few channels, which once again is favourable for the localisation of economic mineralisation.

Fluid mixing can be an important additional mineralising process at this level also.



Figure 10:Boiling point for depth curve

2.6 Applications of These Principles in Exploration

We now have a picture of what the target zones for epithermal mineralisation might look like: at depth they will be steeply-dipping structures, with economic mineralisation concentrated on a few major channels. At shallower levels they may flare into stockworks, within which, while individual veins will still be predominantly vertical, the zone as a whole might be more laterally extensive. But the need for permeability to be constantly rejuvenated means that the economic zones will be concentrated on major structural channels. This will especially apply to any lateral outflow zones, simply because the available fluid is potentially spread over a much wider (radial) area.


0

200

400

600

800

1000

1200

1400

1600

100 150 200 250 300 350

BPD.xls

Temperature (°C)

Dep

th (

m)

k:geo\lec\min\98min\mod1\figbpd2.ppt


Thus structural studies should be an important part of the exploration programme. These should include detailed fracture and vein mapping in the field, in the course of detailed prospecting, trenching and so on. But it is also important to consider structures on the larger scale. This may be best done using aerial photography or satellite imagery, especially in steep or jungle-covered terrain. Side-scanning radar imagery can also be very useful, especially in areas without good topographic maps. It is better than aerial photography for revealing the fine detail of topography in heavily vegetated areas. It is best to use the different types of image in combination.

Using these techniques it may be possible to identify target structural zones which can then be investigated in detail. It is important to appreciate the scale of such structures. Experience in active systems shows that major faults can often be traced over 1 to 5 km, and sometimes three times as far. Stratigraphic mapping should not be neglected, as significant stratigraphic dislocation may imply the presence of a major fault. But in volcanic areas, erection of a sufficiently detailed and meaningful stratigraphy may be impossible. Fluid inclusion and alteration studies can reveal faulting on a smaller scale.

It is equally important to avoid mapping spurious "faults". Particularly where there is little exposure and the geologist finds it difficult to explain the observed distribution of mineralisation, there can be a tendency to assign all observable surface lineations to faults. Without direct evidence such as sheared or slickensided outcrops, inconsistent juxtaposition of lithologies, or strong lineation of fossil thermal activity and/or alteration, a fault should only be inferred if other explanations can be discounted. Lineations that can be mistaken for faults include:

Bedding planes or foliation.

Radial drainage off conical volcanoes (Figures 11, 12)

Raised coastal or alluvial terraces.

Junctions between lava flows or lahar surfaces.

Similar comments apply to the interpretation of ring-faults and calderas. Such structures do occur in volcanic areas and in some case can constitute significant permeable targets for exploration. But not all sub-circular structures are of this origin. Other explanations include slump amphitheatres, lithological contacts at the margins of a volcanic ring-plain (Figure 13), eroded structural domes and large-scale karst features.



Figure 11: Development of lineations during geomorphic development


(a) Examples of lineations without structural significancedeveloped during erosion of conical volcanoes

(b) Apparent lineation created by sub-parallel dendritic drainage on outflow fan.


Lavasand

lahars

Opposed stream

s

Lith

ostr

atig

rap

hic

con

tact

Adjacent streams


Figure 12: Stream piracy producing non-structural lineation



(b) Following capture

(a) Initial drainage


Figure 13: Development of non-structural arcuate drainage


(a) Initial cone

(b) Later flows divert drainage

(c) Development of concentric arcuate streams

Drainage diverted

Concentric arcuate streams

K:geo\lec\min\99min\mod15\fig11.ppt

M03 Permeability Controls

Documents

Transcript of M03 Permeability Controls