Mechanics of low-angle normal faulting: An example from Roosevelt Hot Springs geothermal area, Utah

Tectonophysies, 86 (1982) 343-361 Elsevier Scientific

Publishing

MECHANICS ROOSEVELT

RONALD

343

Company,

Amsterdam-Printed

in The Netherlands

OF LOW-ANGLE NORMAL FAULTING: AN EXAMPLE HOT SPRINGS GEOTHERMAL AREA, UTAH

L. BRUHN,

MICHAEL

R. YUSAS and FERNANDO

FROM

HUERTAS

Department of Geology and Geophysics, University of Utah, Salt Lake City, Utah 84112 (U.S.A.) (Received

June 3, 1981; revised version accepted

December

23, 1981)

ABSTRACT

Bruhn,

M.R. and Huertas,

R.L., Yusas,

from Roosevelt Roosevelt geothermal

Hot Springs

dippin

ranged

of faults and joints

rocks.

Low-angle

denudation

faults,

westward

of the reservoir’s dipping

The average

between

altered

conditions

Hydrothermal

alteration

in the exhumed

denudation

the structqral

planes

of sliding

friction

stability

possible

faults

an example

of southern

of weakness along

along

the faults

simultaneously

The an

with steeply

faults at depth.

which

the denudation

faults

was less than 0.5 and probably

of the denudation

in the fault zones indicates reactive

Utah.

5’ and 35” to the west, form

devetoped

for formation

of chemically

faults, indicating

faults was estimated that Faulting occurred

fluids. The hydrothermal

in the present

geothermal

that the frictional

as a consequence

effects of this hydrothermal

reservoir

resistance

lower than along similar structures

of the reservoir

mechanical

depth

preserved

faulting:

alteration

may

in the fault zones.

along fractures

could be significantly

Mountains

fault plates and merge into denudation

cataclasite

friction

normal

in Cenozoic plutonic and Precambrian

between

These

provided

in the presence

reduced

dipping

structure.

0.15 and 0.4. The maximum

have significantly

reservoir

joints

coefficient

as 5 km. Hydrothe~ally brittle

in the Mineral

by systems

faults that dissect the low-angle

nucleated.

consider

area is located

is formed

component

Gently

under

geothermal

of low-angle

Tectonuphysics, 86: 343-361.

area, Utah.

reservoir

metamorphic important

F., 1982. Mechanics

Hot Springs geothermal

is similar to that observed

along faults and joints

in unaltered

of fluid withdrawal

granitic and

in the

rock. Studies

reinjection

of

should

alteration.

INTRODUCTION

Low-angle normal faulting is commonly associated with Cenozoic- deformation in the Basin and Range Province of the western United States (Armstrong, 1971; Davis and Coney, 1979). The faults may develop in either brittle or ductile environments of deformation depending on their depth in the crust and local geothermal gradients. Denudation faults form an important type of low-angle normal fault associated with the movement of cover rocks from uplifting uplands toward adjacent valleys or grabens. Faulting results in extensive brittle deformation in the fault plates with 0040- 195 I /82/oooO-0000/$02.75

0 I982 Elsevier Scientific

Publishing

Company

144

cataclasis,

jointing

and the development

into gently dipping Information solving

about

problems

faulting

understanding

is a fundamental

leum reservoirs

problem

Knowledge economic

rift provinces, the mechanics

in regional

brittle

is important

such as the Basin and Range

tectonics,

dipping

Also geothermal

seismicity faulting

deformation

to

Denudation

behavior

and reservoir

is initiated

normal

and petro-

fault plates and the frictional

at which denudation

The intense

faults

engineering.

of these gently

affect production-induced

of the depth

implications.

faults that often merge

of denudation

as well as reservoir

may exist in denudation

the faults could strongly tion experiments.

normal

at depth.

behavior

tectonics

in volcano-Teutonic

Therefore,

of high-ail~le

surfaces

the frictional

in regional

is common

Province. faults

decollement

of

stimula-

has important

in the fault plates is ideal for

the development of fractured reservoirs that may contain commercial quantities of petroleum or geothermai fluids (Kostura and Ravenscroft, 1977). The potential volume of a reservoir will be directly related to the maximum depth of low-angle faulting. Recent advances in our knowledge of rock properties and the magnitudes of stress in the crust provide new impetus to investigate the mechanics of low-angle normal faulting. An extensive amount of information is now available from laboratory testing

concerning

rock properties

under

brittle

conditions

and,

in particular,

the

behavior of anisotropic rocks containing crack arrays, joints and faults (Paterson. 1978; Logan, 1979). Furthermore, the results of field studies in part of the Basin and Range

Province

geological

suggest

environments

that laboratory

data

can be successfully

in the upper crust (Zoback

and Zoback,

extrapolated

to

1980). Finally,

the

results of many stress measurements allow one to estimate stress gradients up to several kilometers (Jamison and Cook, 1980; McGarr, 1980). In this paper we integrate theoretical

rock

mechanics

information to investigate

from field geology, laboratory the

frictional

properties

at depths testing and

of low-angle

normal faults in the Mineral Mountains of southern Utah. The terrane contains a commercially exploitable geothermal area, therefore our study has implications for geothermal

reservoir

properties

as well as tectonics.

FAULT SYSTEM AT ROOSEVELT HOT SPRINGS GEOTHERMAL

AREA

Roosevelt Hot Springs geothermal area is located in the west-central flank of the Mineral Mountains in southern Utah (Fig. 1). The geothermal reservoir is formed by a complex system of faults and joints that developed in the Precambrian and Cenozoic crystalline rocks after consolidation of the Tertiary plutonic complex (Lenzer et al., 1976). The faults and joints evolved simultaneously. Sets of systematically oriented joints formed in response to thermo-mechanical stresses during uplift and cooling. However, the intensity of jointing was partly controlled by faulting, as joint intensity increases dramatically next to fault zones. Conversely, the attitudes of

LEGEND @,I =

QUATERNARY

‘v = RHYOLITE FLOWS PYROCLASTICS v = RHYOLITE P = PRECAMBRIAN 8

DOMES

TERTIARY

/FAULTS,

8

FELSIC DASHED

CONCEALED

/I

:‘400

OR

mw In-2

PLUTONS WHERE INFERRED

HEATFLOW

CONTOUR

GNEISS

SCHIST

Fig. 1. Location Geology

0

DEPOSITS

map of Mineral

Mountains

and geologic

after Evans and Nash 91978) and Nielson

map of Roosevelt

Hot Springs geothermal

area.

et al. (1978).

faults were partly controlled by the joints. For example, denudation faults apparently formed along a gently westward dipping joint set. Low- to moderate-angle denudation faults occur throughout the Mineral Mountains, but are most common in the geothermal area along the west-central flank of the range (Sibbett and Nielson, 1980). The faulting post-dated granitic dikes intruded 12-13 m.y. ago (Evans, 1980) and pre-dated the extrusion of rhyolitic domes and flows dated at 0.4-0.5 m.y. (Evans and Nash, 1978). Most denudation faults in the geothermal area dip toward the west between 5O and 35” in Precambrian metamorphic and Tertiary plutonic rock (Fig. 2). Some of these faults have listric geometries with the dip steepening up-section to as much as 65”. The denuation fault plates are cut by numerous northwest-trending faults that dip between 70” and 90’ (Figs. 1 and 2). Many of these faults root in denudation fault zones at depth, proving that movements on the high and low-angle faults occurred simultaneously. The amount of displacement on the faults is generally difficult to determine due to the lack of marker units in the crystalline rock. The geometry of the denudation

Oal-QUATERNARY Tp

DEPOSITS,

-TERTIARY

FELSIC

Pg -PRECAMBRIAN ,ZX-

f-

GNEISS

LOW-ANGLE INDICATE MAJOR

CATACL INFERRED

FAULTS,

MOSTLY

ALLUVIUM

0

I Km

PLUTONS 8

SEA

ASTK

ZONES,

SENSE

OF DISPLACEMENT

ARROWS

LEVEL

SCHIST

INDKATE

ARROWS

SENSE

OF

DISPLACEMENT

Fig. 2. East--west

cross section across part of Roosevelt

1978). Small cataclastic

zones are schematic.

Hot Springs

geothermal

The line of section is indicated

area (after Nielson et al.,

in Fig.

I.

faults and the associated ~~-angle fault set indicates that movement was toward the west or southwest. This direction of movement is also indicated by slickenside orientations in many of the smaller fault zones. Nielson et al. (1978) estimated 6 10 m of displacement in a S80°W direction from offset lithologies on part of a large denudation fault at a locality just south of the geothermal area. Displacements on many of the smaller faults, formed by the structural reactivation of individual joints are likely to be several meters or less. The faults contain hydrothermally altered cataclastite in zones up to 15 m thick indicating that deformation proceeded under brittle conditions. The altered cataclasite consists of cornminuted quartz and feldspar grains surrounded by newly grown sericite, chlorite, epidote, and clay minerals. The alteration assemblage is typical of low-grade indicate

greenschist alteration

facies mineralization. temperatures

between

Fluid-inclusion 200°C

studies of the cataclasite

and 3OO“C (C. Clark,

pers. com-

mun., 1981). Chlorite and sericite form lineated fibrous growths on micro-fault surfaces in the fault zones, proving that hydrothermal alteration occurred during faulting. In some areas, micro-diorite and rhyolite dikes were injected along the faults and subsequently deformed as the result of renewed movement (Nielson et al., 1978; Yusas and Bruhn, 1979). The youngest faults in the geothermal area consist of normal faults that strike north-northeast and cut Quaternary deposits (Fig. 1; Petersen, 1975). These faults post-date the high-angle faults that formed denudation faulting. Another set of faults trends eastward, cutting across the strike of the mountain range. Little is known about the history of this latter fault set, although these vertical, east trending faults may form important structures in the geothermal reservoir (Ward et al., 1978). Joints represent an important component of the reservoir’s structure and provide

347

passageways relatively

for water infiltrating

homogeneous

of three major joint extension

joints

pe~endicular

the central

Mineral

system.

Mountains;

The joint

system

generally

consisting

sets (Figs. 3 and 4). Two sets of steeply dipping,

trend

northward

and

eastward,

to the strike of the contacts

Joint spacing parts

through

into the geothermal

is variable,

of the plutonic

ranging

between

occurring

roughly

the igneous

sub-orthogonal parallel

and country

from 1 m to 30 m in the more structurally

complex

to less than

a centimeter

in areas

is

where

and rocks.

coherent intense

faulting has occurred. In these latter areas, many of the joints have undergone shear displacements and now consist of hydrothermally altered, anastomosing fractures with slickensided surfaces. The third major joint set consists of gently to moderately westward dipping joints that generally have smooth, planar surfaces (Figs. 3 and 4). Joint spacing ranges from larger than 1 m in unfaulted parts of the plutonic complex to as little as 10 cm in highly faulted regions. In these latter areas, the joints sub-parallel the denudation faults and many of them contain hydrothermally altered cataclasite formed by shearing.

-FIELDS OFAVERAGE JOINT PLANE GEOTHERMAL AREA (3 DOMAINS) (,=‘

POLES

- FIELDS OF AVERAGE JOINT PLANE PLUTON’S INTERIOR (3DOMAlNS) S

- JOINT

SETS

Fig. 3. Summary contains

CONTAIN

diagram

stereographic

of average

that are reported

projection.

POLES

IN

CATAC&AStTE

the average joint orientations

2ooO measurements

IN

joint

orientations

for several domains

in the central

Mineral

of similar structure.

in Yusas and Bruhn (1979)..Data

Each

field

The data represents

Mountains.

over

are plotted on a lower hemisphere,

FRACTURE ROOSEVELT

Fig. 4. Cartoon geothermal

showing

SYSTEM

IN

GEOTHERMAl_

the typical joint

area. Note the increased

AREA

and mesoscopic

intensity

of fracturing

fault geometry adjacent

in the Roosevelt

Hot Springs

to the faults.

It is important to note that #he joint system in the Precambrian rocks is similar to that in the pluton. This indicates that the ancient and complex structural fabrics in the metamorphic rocks did not control jointing during uplift of the terrane in the Tertiary. We determined the micro-crack fabric in quartz grains from a number of samples coilected in the plutonic complex and Precambrian country rock. The fabrics are orthorhombic and consistently oriented between the various sites. There is good correlation between the orientations of micro-cracks and joints in outcrops from which the samples were collected (Fig. 5) Orthorhombi~ crack fabrics are often found in crystalline rocks and presumably reflect the effects of applied and residuai stress release during uplift and decompression (Schdtz and Koczynski, 1979). Notably, principal residual stress orientations in outcrops of Mineral Mountains’ granite and gneiss generally correspond with joint orientations, and, to a lesser degree, with the micro-crack orientations (Fig. 5). MECHANICS

OF LOW-ANGLE

NORMAL

FAULTING

The Late Cenozoic stress field in the Mineral Mount~ns has been extensional with a verticai maximum principal compressive stress (c~) and a northwest to

349

OLES

99

POLES

Fig. 5. Relations between the orientations of joints and microfractures in granite and gneiss outcrops. The data are plotted on Iower hemisphere stereographic projections and contoured by the Schmidt method. Contour intervals in percent per 1Warea are: (I, c, e = 1, 2, 3 and 5%, b and d = 1,4, 8 and 12% andf = 1, 5, 9, and 15%. The maximum horizontat extension axes determined from in-situ residud stress tests are indicated by the large arrows. The original data may be found in Yusas and Bruhn (I979).

stress (o-0. The mountain

southwest-trending

minimum

principal

is located

transition

zone

between

the

(Fig. I). Regional

extension

in this part of the transition

in the

Colorado

Plateau

have begun

compressive

Basin

and

as early as 26 m.y. ago and was certainly

underway

(Rowley

et al., 1978). The onset of extension

was marked

regional

talc-alkaline

basalt-rhyolite

volcanism

to bi-modal

by the formation of north-trending which indicates a regionally vertical

Range

range

Province

and

zone may

by 20 m.y. ago

by an abrupt volcanism

change

from

accompanied

horsts and grabens. This style of tectonism, 0,. has continued to the present.

The orientation of igneous dikes, faults and aligned rhyolite domes in the Mineral Mountains indicate that since at least 12 m.y. ago the local stress field was similar in orientation to the regional field (TableI). This means that the dei~udation faults apparently formed at angles in excess of 45’ to a,; at much larger angles than are commonly observed in rock deformation experiments (Paterson, 1978) and field studies (Hobbs et al.. 1976). The angle between

paleo- u, and an individual

fault can be determined

from the

attitudes of extension fractures in the wall rock. Specifically, (I, was oriented down the dip of the pervasive north-trending extension fractures that occur next to faults while us was perpendicular to them. We have studied the extension fractures at several locales where a denudation fault is exposed. In general, the extension fractures in the wall rock dip between 70” and 9O”W, regardless of the dip of the fault. This indicates that the angle between u, and the segments of denudation faults dipping

TABLE

loo-2O”W

was as large as 70”-80’.

I

Paleo-stress

field at Roosevelt

Hot Springs Geothermal

Resource

Geologic

evidence

Time (m.y.)

Stress orientation

Quaternary

e, -vertical

N-NE

striking

o,-W-WNW

normal

faults in alluvium

q-E-W

rhyolite

domes aligned

6, -vertical

high and low-angle

a,-W-WSW

faults striking

CI,-vertical

steeply dipping

a,-W-WNW

dikes striking

0.4-

0.5 **

Denudation

faulting

9 ***

12 -13 ***

* e, = maximum ** K/Ar *** K/Ar

principal

*

Area

et -vertical

steeply dipping

I+-W-WNW

dikes striking

compressive

press, es --minimum

dates from Evans and Nash (1978). dates from Evans (1980).

principal

normal

N-NNW rhyolite northward granitic northward

compressive

stress.

351

Rock properties

Maximum values for the coefficients of internal friction and strengths of Mineral Mountains granitic rock can be estimated from available field and laboratory data. The Quaternary faults along the western flank of the range dip between 70” and 75’ at the surface (Fig. 2). This dip angle corresponds to internal friction coefficients (pi) between 1.2 and 1.7 for the igneous and metamorphic rocks cut by the faults. A failure envelope for unaltered Mineral Mountains granite is presented in Fig. 6. The envelope is constructed from the results of three tri-axial compression tests by Pratt and Simonson (1976) and tensile strength tests by Yusas and Bruhn (1979). The cohesive strength (C) is 20 MPa and the coefficient of internal friction (pi) ranges between 1.6 and 0.6 with increasing confining pressure. The uniaxial compressive strength (S,) is 121 MPa. A tensile strength of 5 * 2.4 MPa was determined from twenty Brazil tests on samples collected by Yusas and Bruhn (1979). The large standard deviation in the test results reflects a strong grain size dependence in tensile strength. Eight of the samples had average grain diameters of 1 mm and an average strength of 6.9 t 1.7 MPa, while the other samples with 2 mm average grain dimensions had a tensile strength of only 2.5 * 0.6 MPa. Presumably, this decrease in strength with increased grain size reflects the presence of longer flaws or cracks in the larger grains (Brace, 1961). Griffith crack theory predicts that the rock with bigger grains should be

MINERAL

MOUNTAINS FAILURE

GRANITE

TEST

#

a3

(MPa)

o;-5

(MPa)

ENVELOPE

‘TV = NORMAL

STRESS

( MPa)

Fig. 6. Failure envelope for unaltered Mineral Mountain’s granite. Compression Simonson (1976)and tension tests by Yusas and Bruhn ( 1979).

tests by Pratt and

weaker;

as is commonly

We assume crystalline rocks. igneous

that

observed

the

failure

rocks in the Mineral

This contention

tests. in

Mountains.

is supported

and metamorphic

on Precambrian

in strength envelope

Fig. 6 is applicable

including

to the various

the Precambrian

by the fact that the Quaternary

rocks have sinular

metamorphic faults

in the

dips and by the results of Brazil tests

gneiss. These tests gave an average tensile strength

of 6.7 -!m0.6 MPa

for four samples with 1 mm average grain diameters; essentially identical strengths to those measured on the igneous rocks with similar grain sizes (Yusas and Bruhn, 1979). No compressive strength tests have been reported for the Precambrian gneisses. It should be noted that most of the igneous and metamorphic rocks in the mountain range are quartzo-feldspathic and individual samples from one phase often overlap in composition with those of other phases (Nielson et al., 1978). This similarity in composition is probably the underlying cause for the relatively homogeneous mechanical

properties

inferred

should

however,

that Nielson

be noted,

from the available

field and laboratory

et al. (1978) found

a greater

data. It

intensity

of

faulting in the Tertiary hornblende gneiss than in other units and the amount cataclasis was much less in the biotite-rich gneisses than in the other rocks.

of

Fault mechanics Hypotheses for the origin of denudation faulting must account for the simultaneous evolution of the denudation faults and the steeply dipping, northwest striking faults in the fault plates (Fig. 2). If u, was vertical, as we have inferred from geological data (Table I), then the steeply dipping faults could be predicted given the failure envelope in Fig. 6. However, the gently dipping denudation faults would not be expected to develop. We suggest that the gently to moderately westward dipping joints and micro-crack which the denudation

arrays in the granitic rocks formed planes of weakness along faults nucleated. This would explain why many of the gently

dipping joints have locally undergone shearing, particularly in the vicinity of major denudation faults. Also, the most probable explanation for faulting at angles greater than 45” to u, is that the rock was anisotropic, with preferentially oriented planes of weakness. Either joints or planar arrays of micro-cracks such a strength anisotropy in crystalline rocks (Walsh Cook, 1976; Paterson, 1978).

are the most likely causes of and Brace, 1964; Jaeger and

Two physical parameters are most likely responsible for the evolution of low-angle normal faults. These are excessive pore fluid pressure or low coefficients of sliding friction. Shearing along faults or joints will be greatly facilitated by excessive pore fluid pressures (Hubbert and Rubey, 1959). However, the process of continued fracturing in the denudation fault plates should have progressively increased the fracture permeability of the rock, and consequently, tended to decrease excessive pore pressure. Consequently, high pore pressure should not be arbitrarily assumed as

353

the mechanism that allowed low-angle normal faulting. The potential effects of rock friction on the mechanics of denudation faulting can be investigated using experimentally derived data on the frictional properties of joints and faults. Coefficients of friction for sliding along pre-cut surfaces in felsic rocks generally vary between 0.5 and 0.8 (Jaeger and Cook, 1976; Paterson, 1978) and between 0.6 and 1.0 for sliding on a~ifi~ally formed shear surfaces (Paterson, 1978). The equation that is most representative for repeated sliding on pre-existing surface is 7S= 0.85 a,, where 7S= shear stress to initiate sliding and cS = effective normal stress across the surface (Byerlee, 1978). This equation holds over large ranges in rock composition, gouge thickness, confining pressure, pore pressure, temperature and strain rate in the laboratory at normal stresses to 2 kbar. If a, 2 2 kbar, pS is estimated as 0.6. Jamison and Cook (1980) have used the results of stress tests in the uppermost crust to estimate the cohesive strengths and friction coefficients of natural faults and joints. They found that the cohesive strength (C) was nearly zero and that coefficients of sliding friction (~$1 ranged between 0.22 and 0.65. The strength of an anisotropic rock varies as a function of the angle (8) between o, and the weakness plane (Jaeger and Cook, 1976). The differential stress needed to fail the rock can be found from the following equation which is derived from the discussion of Jaeger and Cook (1976, pp. 25-26).

In this equation C is the cohesive strength and cS the coefficient of sliding friction for the plane of weakness; B is the angle between e, and the perpendicular to the plane. If ui is vertical, as assumed in this study, then 0 is also the dip angle of the plane of weakness. The rock strength given by eq. 1 can be normalized with respect to the strength given by the failure envelope (Fig. 6) for rock lacking a throu~goi~g anisotropy plane and plotted against 8 to produce the curves shown in Figure7, These curves represent several different values for the friction coefficient (it,) on the weakness plane and show how the normalized strength of the rock (Au*) varies as a function of the dip angle of the anisotropy plane (a, is assumed to be vertical). A graph like that in Fig. 7 can be constructed for any depth of interest if the vertical shear stress gradients are known. McGarr (1980) reports maximum shear stress (r,,) variations with depth in crystalline rocks based on hydrofracture stress tests. The equation rmax.=A + BZ can be fitted to hydrofractu~g data, where A and 3 are constants and 2 is depth in kilometers. For fifteen measurements in extensional regions A = 1.00 f 3.76 MPa and B = 6.69 -C 1.93 MPa/km. Consequently, we assume that the paleo-shear stress in the Mineral Mountains can be estimated from the equation T,,,= = 1 MPa + 7 MPa,/km. Notice that Fig. 7 has been constructed for a depth of 2 km. A pore pressure equal to 125% of the hydrostatic

DEPTH=

2

Km.

FLUID PRESSURE= HYDROSTATIC COHESION rmax

1

j

IIIII

O0

I

IO’

ZOO 30°

DIP

Fig. 7. Estimated

OF

and friction

divided

by the strength

I

I

I

I

I

40°

50°

60’

70’

00’

ANISOTROPY

strength

cohesion represented

= 0

I MPa+7MPa/Km

I

, 0

=

125%

PLANE

of Mineral

coefficients

Mountain

90° (81

granitic

rock containing

between 0.1 and 0.5. Ao* is the strength

of intact rock that lacks a through-going

by the failure envelope

anisotropy ((T, -a,)

plane of weakness.

in Fig. 6. The Figure is discussed

planes

with zero

of the anisotropy

plane

The latter strength

is

in the text.

value is required to cause high and low-angle faulting simultaneously. The curves in Fig. 7 can be used to determine the minimum dip angle (e,,,,,) for an anisotropy plane along which sliding can occur given a specific coefficient of friction. strength

This is due to the fact that u, was vertical. The intersection of the U-shaped curve with the line Au * = 1 lies directly above the minimum possible dip

angle (19,,)

for a low-angle

fault. For example,

6,,,i, in Fig. 7 is 5.5’, 14’ and 22” for

faults with friction

coefficients

(CL,) of 0.1, 0.3 and 0.5 respectively.

We constructed friction coefficients

graphs like Fig. 7 for depths between 2 km and 5.5 km with ranging between 0.1 and 0.6. In each case we determined S,,,,,

and listed the results in Table 11. The minimum depth of 2 km was selected because the known thicknesses of preserved fault plates is just less than that depth. The pore pressure geometry

needed to initiate high-angle faulting at any depth is limited by the of the laboratory failure envelope and the principal stress difference, which

is fixed by the assumed shear stress gradient. Inspection of Table II shows that pS 2 0.5 is too large, because there is field evidence that the angle between u, and some of the larger denudation faults was 70’ or greater, requiring that @,,., G 20”. An upper bound for the friction coefficient must be in the range pS < 0.5. This upper bound is much lower than the commonly cited values of 0.6 found for sliding on faults in the laboratory (Byerlee, 1978) but is within the range of friction coefficients estimated from stress tests (Jamison and Cook,

1980).

355

TABLE

II

Minimum faulting

dip angle

for denudation

FRICTION 0.1

* Pore

faults

as a function

of the coefficient

of friction

and

depth

of

*

pressure

0.2

required

COEFFICIENT 0.3

0.4

to cause

(4) 0.6

0.5

denudation

faulting

is about

125% hydrostatic

if T,,,~ =

I MPa+

7 MPa/km.

We have attempted to estimate a lower bound for the friction coefficient by calculating the resisting 1forces acting on one of the preserved denudation fault plates (Fig. 2). The two-dimensional geometry and dimensions of the plate (Fig. 8) are taken

from cross section

B-B' of Nielson et al. (1978). The basal denudation

has been offset along younger,

high-angle

faults, but the numerous

low-angle

fault joints

in the fault plate probably provide potential slip-surfaces connecting the larger segments of the old fault. Therefore, the denudation fault should still be capable of movement. The subsurface structure along the western margin of the mountain range is from the seismic refraction work of Gertson and Smith (1979). The pore pressure in the geothermal field is presently hydrostatic. The method

of analysis

is a modification

of Janbu’s

calculating the factor of safety on a non-circular terms into Janbu’s original equations to account

(1954) “method

of slices” for

fault. We have introduced for the effects of tectonic

new shear

stresses and resisting forces due to abutting of the fault plate against valley fill (Appendix 1). The coefficient of friction is found by setting the gravitational and tectonic driving forces equal to the resisting forces caused by the weight of the valley fill and frictional resistance along the non-circular fault surface. This assumes that the fault plate is in static equilibrium because there is no geological evidence for contemporary movement on the low-angle fault. Satisfying the assumption provides a lower bound estimate for the frictional coeficient (p,), if the cohesion of the fault is negligible. The calculation suggests that ps = 0.3 is a reasonable estimate for a lower bound on the coefficient of friction. If tectonic stresses are not considered.

Fig. 8. Fault plate geometry discussed

TABLE

the lower bound

Coefficient

Tmar.

estimate

of friction = 1 MPa+

for static coefficient

of friction

(a,).

The figure is

of friction Ratio of resisting

(p,)

0.2

0.2

0.6

0.3

0.8

0.4

1.1

0.5

1.3

Cuse 2: We set tectonic

stresses equal to zero and consider

I

forces *

only body forces

0.5

0.15

0.9

0.2

1.3

* The ratio of resisting is greater

to driving

7 MPa/km

0.1

0.

coefficient

1.

III

Lower bound

Cuse I :

used to calculate

in the text and Appendix

to driving

force equals one if the fault is in static equilibrium.

than 0.3 while for cure 2 ps is greater

than 0.

IS.

Thus. for cuse

1 ps

357

p”G, = 0.15 (Table III). We suggest, then, that the coefficient of sliding friction along the denudation faults was on the average, less than 0.5 and probably greater than 0.15. The data in Table II can be used to estimate the maximum depth for denudation faulting. Unfortunately, @,, is a slowly varying function of depth; therefore, the “maximum depth of faulting” estimate is poorly constrained. Large segments of the denudation faults apparently formed at angles between 70” and 80” to a,; this corresponds to 0,, as small as 10”. If the coefficient of sliding friction was greater than 0.15, then the maximum depth of faulting was constrained to depths at or above 5 km (Table II). DISCUSSION

We suggest that the flow of hydrothermal fluids along low-angle joint surfaces and crack arrays led to the nucleation of denudation faults in the extensional stress field. Subsequent movement along the faults was then sustained by the continued effects of fluid flow. The mineralogy of the cataclasite in the low-angle fault zones is different than that of the country rock granite and gneiss. The cataclasite contains a hydro~ermal alteration mineral assemblage that includes chlorite, sericite, epidote and clay minerals. Fluid inclusions in quartz indicate fluid temperatures above 2OOOCduring alteration (C. Clark, pers. commun., 1981) and inspection of thin sections indicates that these newly formed minerals partly replaced the original mineral grains in the country rock. Notably, modal feldspar is drastically reduced in the cataclastic relative to the original rock. There is good textural evidence for continued alteration during faulting and locally, mesoscopic fault surfaces are coated with fibrous quartz and phy~osi~~tes indicating that pressure solution slip was a subsidiary process to cataclasis. Hydrothermal alteration may facilitate faulting in several ways. First, the presence of a chemically reactive fluid in the fault zone may lower the rock’s strength through hydrolization of silicate bonds (Scholz, 1968) and by reducing the amount of highly brittle minerals such as feldspar by recrystallization to new phases (Mitra, 1978). Secondly, the formation of phyllosilicates could decrease permeability in the fault zone and cause localized zones of high fluid pressure (Wang, 1980). Also, clay minerals such as vermiculite and montmorillonite have interlayer water that can cause low coefficients of friction due to a pseudo-pore pressure effect (Wang and Mao, 1979). Further work on the relative importance of the various chemical and physical effects of the alteration in enhancing faulting is in progress. The friction estimates of p, ( 0.5 have potentially important implications for exploitation of the present geothermal system. Hydrothermal alteration along joints and faults in the geothermal reservoir is ~ntinuing at the present and the mineral assemblages are similar to those found in the exhumed denudation faults (Parr-y et al., 1978). This implies that the resistance to sliding on joints and faults in the

reservoir

might

be considerably

rock. A thorough for engineering large-scale

investigation studies

lower than

of the potential

fluid extraction

in the surrounding

of the mechanical

unaltered

effects of this alteration

structural

stability

granitic is

of the reservoir

needed during

and re-injection.

Nielson et al. (1978) suggested that plates of crystalline rock could have been faulted over valley fill along some of the denudation faults. We have not taken topographic stresses into account in this study since the paleo-topography of the mountain range and its history of uplift are not known with any certainty. However, the shallow relatively

depths

at which

low coefficients

hypothesis

denudation

faulting

of sliding friction

is mechanically

is expected

to occur

that we have estimated

and

the

suggest that their

feasible.

CONCLUSIONS

Denudation gently dipping

faults in the Mineral Mountains developed by shearing along a set of joints and cracks. The average coefficient of friction along the faults

was less than 0.5 and probably in the range of 0.15-0.4. The estimated friction coefficients are lower than the commonly cited values of ~~“0.6 determined in laboratory faulting tests. However, our estimates are similar to the range of values cited by Jam&on and Cook (1980) for friction on natural faults and joints in the uppermost

crust (0.22 d ps < 0.65). The maximum

is estimated

as 5 km, although

Hydrothermal suggest

that

the

this estimate

alteration

is ubiquitous

alteration

was

depth

for formation

of the faults

is poorly constrained. in the denudation

necessary

for

faulting

fault

to have

zones

and

occurred.

we Such

alteration is common in many fault zones and much work needs to be done investigating the effects of chemically reactive fluids on rock friction.

in

ACKNOWLEDGEMENTS

We thank D.L. Nielson and R.B. Smith for reviewing a preliminary draft of this manuscript. The work was supported by the United States Department of Energy under

contract

Corporation APPENDIX

with

a grant

from

the

Research

I

An estimate gently

of the lower bound

for the static coefficient

of the large denudation

dipping

joints

segments

of the fault

assumed

to be hydrostatic

equation

rmax =

horizontal

and

to R.L. Bruhn.

the cross-section many

DOE/DGE-ET/28392-31

where

in the fault it is off-set

components

is found by assuming

plate

of weakness connecting the faults. The fluid pressure is

provide

by younger,

(McGarr,

shear

potential high-angle

stress

at any given depth

1980). The resisting

and set equal by adjusting

planes normal

the magnitude

and driving of the friction

Presumably,

that

equilibrium.

and the maximum

1 MPa+ 7 MPa/km

of sliding friction

fault plate in Fig. 2 is in static

is estimated

from

forces are resolved coefficient

the

the into

(p,) on the

359

fault using the non-circular account

for tectonic

fault analyses

of Janbu

The fault plate is divided into six segments The weight per unit volume 22.5 kN/m3. Jamison

of crystalline

The cohesive

6

strength

as 26.5 kN/m3

and that of the alluvium

to be negligible,

following

as

the results

of

N,

=

6

+R=l//,

resisting force

pighi; pi =density

z W,tar1cr,+7~sin2a~Ax~ ,=I drivmg force

and g = [gravitational

u, = w, =

weight of segment

ai

=

average

1

maximum

shear stress at depth “hi”

correction

factor

“i”.

dip angle of segment

tana,)

“i”. MPa/km.

forces. This factor is found by taking the ratio A/L.

fig. I14 of Hoek and Bray (1977). Here, f. = 1.013.

cos2ai.

force due to valley fill abutting

zr

found from r,,,_ = I MPa+7

to allow for inter-segment

(Fig. 8) and consulting (l+ps

accelerationl.

at depth hi.

pore pressure

= =

against

the western

end of the fault plate (Fig. 8).

l/2 W,Z*, where W, = average weight per unit volume of the Tertiary (-23.5)

kN/m3.

The tectonic and

of mean height (hi), length ( Axi), and unit width (Fig. 8).

rock is estimated

of the fault is considered

1

(Pi-Li)Axi

’ ,=,

N, R R

to

to be solved for ts is:

Ps

;;

equation

and Cook (1980).

The equation

P,

(1954). We have added a term to Janbu’s

shear stress on the fault plane.

resolving

compressive Axi/cos

force term (7i sin 2ai Axi) for each segment it to find

the shear

stress is assumed

ai. The horizontal

The remainder

stress

on the fault

to be vertical.

component

of the equation

results are presented

and Quaternary

valley fill

The depth (Z) to the fault is 1070 m. is found by calculating

dipping

The total shear

at angle

force on the fault

of this force is (7i sin 2ai Axi/cos

is derived

and discussed

by Janbu

7i for each segment

ni. The

maximum

segment

ai)(cos

ai)=

pricipal

is 7i sin 2cr,

7, sin 2ai Axi.

(1954) and Hoek and Bray (1977). The

in Table III.

REFERENCES

Armstrong,

R.L.,

Nevada

1971. Low-angle

and western

(denudation)

faults,

hinterland

of the Sevier Orogenic

Belt, eastern

Utah. Geol. Sot. Am. Bull., 83: 1729-1754.

Brace, W.F., 1961. Dependence

of fracture

strength

of rocks on grain size. Proc. 4th Symp. Rock Mech.,

Penn. State Univ. Min. Ind. Exp. Sta. Bull., 76: 99-103. Byerlee, J.D., 1978. Friction Condie,

of rocks. Pure. Appl. Geophys.,

K.C., 1960. Petrogenesis

of the Mineral

116: 615-626.

Range Pluton,

Southwestern

Utah. M.S. Thesis, Univ. of

Utah, 92 pp. Davis, G.H. and Coney,

P.J., 1979. Geologic

development

of the metamorphic

core complexes.

Geology,

7: 120-124. Earll, F.N., Utah, Evans,

1957. Geology

of the central

Mineral

Range,

Beaver County,

Utah,

Ph.D. Thesis,

Univ. of

I I2 pp. S.H.,

Jr.,

1980.

Summary

DE-AC07-78ET/28392,

of potassium/argon

University

of Utah, Department

age dating-1979. of Geology

DOE/DGE

and Geophysics,

Rep.

Contr.

Salt Lake City,

Utah, 23 pp. Evans, Rep.

S.H., Jr. and Brown,

F.H.,

, Contr. DE-AC07-80IDl2077,

Lake City, Utah, 23 pp.

1980. Summary University

of potassium/argon of Utah, Department

age datingof Geology

1980. DOE/DGE and Geophysics,

Salt

360

Evans. S.H.. Jr. and Nash, DOE/DGE physics. Gertson.

EY-76-S-07-1701.

rhyolite

from the Mineral

University

of Utah.

Mountams.

Department

C:tah. U.S.A.

of Geology

and Geo-

Salt Lake City. Utah.

R.C. and Smith.

Hot Springs.

R.B.. 1979. Interpretation

Utah and vicinity.

Department Hoeck.

W.P.. 1978. Quaternary

Rep.. Contr.

of Geology

DOE/DGE

and Geophysics.

of a seismic refraction

Rep.. Contr.

Salt Lake City. Utah.

E. and Bray. J.W.. 1977. Rock Slope Engineering.

profile

across

DE-ACO7-78ET28392.

the Roosevelt

IJnivcryitv

of Utah,

i 16 pp.

Institution

of Mimng and Metallurgy.

London.

401 pp. Hobbs,

B.E.. Means, W.D. and Williams,

P.F.. 1976. An Outline of Structural

Geology,

Wiley. New York,

571 pp. Horn,

H.M. and Deere, D.V., 1962. Frictional

Hubbert,

M.K. and Rubey.

characteristics

of minerals,

W.W., 1959. Role of fluid pressure

Geotechmque.

in mechanics

12: 319-335.

of overthrust

faulting.

Geol.

Sot. Am. Bull., 70: 1155166. Jaeger,

J.C. and Cook, N.G.W..

PP. Jamison.

D.B. and Cook,

crust. J. Geophys. Janbu,

N.G.W.,

Kastura.

J.R.

Lenzer,

and

of composite

Ravenscroft,

R.C., Cosby.

Press. New York. 584

values for the state of stress in the Earth’s

slip circles for stability

J.H.,

analysis.

Proc. European

Conference

on

3: 43-49. 1977. Fracture

Controlled

Production.

Am.

Assoc.

Pet. Geol..

G.N. and Berge, C.W.. 1976. Geothermal

exploration

of the Roosevelt

Hot Springs

Proc. Int. Sot. Rock Mech, 3Bl-1.

Liese. H.C.,

1957. Geology

Thesis, Univ. of Utah,

of the Northern

Mineral

Range,

Millard

and Beaver Counties,

Utah.

M.S.

88 pp. (unpublished).

J.M., 1979. Brittle phenomena.

McGarr,

Halsted

Ser., No. 21: 221 pp.

KGRA.

Logan,

1980. Note on measured

of Earth Slopes, Stockholm,

Reprint

of Rock Mechanics.

Res., 85: 1833-1838.

N., 1954. Application

stability

1976. Fundamentals

A., 1980. Some constraints

Rev. Geophys.

Space Phys., 17: 1121-l

132.

on shear stress in the crust from observations

and theory. J. Geophys.

Res., 85: 623 l-6238. McGarr,

A. and Gay.

N.C..

1978. State of stress in the earth’s

crust.

Annu.

Rev. Earth

Planet.

Sci., 6:

processes

involved

in the

4055436. Mitra,

G., 1978. Ductile

deformation Nielson,

deformation

of crystalline

D.L., Sibbett,

of Roosevelt

zones and mylonites:

basement

B.S., McKinney,

Hot Springs

D.B., Hulen, J.B., Moore, J.N. and Samberg.

KGRA,

Beaver County,

Earth Science Lab., Salt Lake City, Utah, Obert,

the mechanical

rocks. Am. J. Sci., 278: 1057-1084. Utah. DOE/DGE

S.M., 1978. Geology

Rep., Contr.

EG-78-C-07-1701.

120 pp.

L. and Duvall, W.I., 1967. Rock Mechanics

and the Design of Rock Structures.

Wiley, New York,

650 pp. Parry.

W.T., Bryant,

J.L.,

N.L.. Dedolph,

1978. Hydrothermal

Rep., Contr.

R.E., Ballantyne,

alteration

ID0/78-1701.a.l.l.,

J.M.. Ballantyne,

at the Roosevelt University

of Utah,

G.H.,

Rohrs,

Hot Springs

thermal

area,

Department

of Geology

The Brittle

Field.

D.T. and Mason. Utah.

DOE/DGE

and Geophysics,

Salt

Lake City, Utah, 29 pp. Paterson,

M.S.,

York, 254 Peterson,

C.A.,

1978. Experimental

Rock

Deformation:

Springer.

Heidelberg-New

pp. 1975. Geology

of the Roosevelt

Hot Springs

area, Beaver County,

Utah.

Utah Geol., 2:

109-l 16. Pratt,

H.R. and Simonson,

Resources Rowley,

P.D.,

between 51-55.

Development Anderson,

the Colorado

E.R., 1976. Geotechnical Agency,

J.J.. Williams, Plateaus

studies

of geothermal

reservoirs.

Report

to Energy

pp. 30-34. P.L. and

Fleck,

and Basin and Range

R.J.,

1978. Age of structural

Province

in southwestern

Utah.

differentation Geology,

6:

361

Scholz,

C.H.,

1968. Microfracturing

and the inelastic

deformation

of rock in compression.

J. Geophys.

Res., 73: 1417-1432. Scholz,

C.H. and Koczynski,

loads. J. Geophys. Sibbett,

B.S. and Nielson,

DOE/DGE

T.A., 1979. Dilatancy

anisotropy

and the response

of rock to large cyclic

Res., 84: 5525-5534. D.L., 1980. Geology

Rep., Contr.

of the central

DE-AC07-78Et28392,

Tullis, J.A., 1979. High temperature

deformation

Mineral

Mountains,

Beaver County,

Utah.

Earth Science Lab., Salt Lake City, Utah, 42 pp. of rocks and minerals.

Rev. Geophys.

Space Phys.,

17:

1137-1154. Walsh, J.B. and Brace, W.F., 1964. A fracture

criterion

for brittle anisotropic

rock. J. Geonhys.

Res., 69:

3449-3456. Wang,

C., 1980. Sediment

Wang,

C. and Mao, N., 1979. Shearing

Geophys. Ward,

subduction

clays

in rock joints

zone

. Geology,

at high confining

S.H., Parry, W.T., Nash, W.P., Sill, W.R., Cook, K.L., Smith, R.B., Chapman,

Roosevelt

Hot Springs

M.R. and Bruhn,

thermal

KGRA.

Geology

and Geophysics,

DOE/ET

M.L. and Zoback,

Geophys.

J.R., 1978. A summary

R.L.,

Springs Zoback,

sliding in a subduction

of saturated

8: 530-533. pressures.

Res. Lett., 6: 825-829.

Whelan, J.A. and Bowman, Yusas,

and frictional

of the geology,

area, Utah. Geophysics,

1979. Structural Rep., Contr.

fabric

geochemistry

D.S., Brown, and geophysics

F.H., of the

43: 1515- 1542.

and in-situ

DE-AC07-78ET28392,

stress analyses University

of the Roosevelt

of Utah,

Department

Hot of

Salt Lake City, Utah, 62 pp. M.D., 1980. Faulting

Res., 85: 275-284.

patterns

in north-central

Nevada

and strength

of crust. J.