CHAPTER

___ 14 ___
INTRODUCTION TO
CROSS-SECTION
BALANCING
Stephen Marshak
Nicholas
Woodward

14-1 INTRODUCTION
Cross sections are very important tools for communicating
information about geologic structures, so the interpretation
depicted on a cross section must be as close to the truth as
possible. The procedure of cross-section balancing has become
popular in recent years as a means of helping to analyze and
improve cross sections. Cross-section balancing permits
geologists to test the validity of the structural geometry
portrayed on a cross section (see Dahlstrom, 1969, and Elliott,
1983). It requires thoughtful analysis of fault shapes, bed
lengths, and cross-sectional areas. One of the key steps
involved in the procedure is the restoration of the beds depicted
on the cross section to the relative positions that they had prior
to deformation.

The purpose of this chapter is to introduce elementary

aspects of cross-section balancing. We provide simple
examples of how a cross section can be constructed, restored,
checked, and improved. Study of cross-section balancing is a
subject that is rapidly advancing, and so the techniques are
constantly being improved. Cross-section balancing is
sometimes an excruciating iterative process that relies heavily
on intuition and on a broad knowledge of structural geology; it
is not suited to a "cookbook" approach. Nevertheless, we hope
that by following the "steps" of our simple examples, you will
grasp the fundamental goals of cross-section balancing and will
be forced to think hard about every line that you draw on a
cross section.
The procedure of cross-section balancing has proven to be
most valuable in the study of deformed belts in which
deformation is largely confined to layers of rock that lie above
a subhorizontal detachment fault or decollement (Rich, 1934;

Rodgers, 1949; 1963). In some literature such belts are called

thin-skinned deformed belts (e.g., Gwinn, 1964; Harris and
Milici, 1977). The thin-skinned concept emphasizes that the
rock below the detachment need not display the folding and
faulting found in rocks above the detachment (Fig. 14-1). The
term thin-skinned was used to describe regions where
deformation was confined to a stratified sequence (cover)
above crystalline basement. If basement was involved in the
deformation, the belt was called "thick-skinned." It is now
known that at some localities detachments lie in crystalline
rock below the basement/cover nonconformity, so the term
thin-skinned is not used as frequently these days. To avoid
confusion, it is best to state simply whether or not basement
rocks are involved in the thrusting.
Deformation involving detachments occurs both in foldthrust belts (e.g., Boyer and Elliott, 1982), in which shortening
of the crust is accommodated by the formation of thrust faults
and associated folds (e.g., Hossack, 1979), and in extensional
or rift terranes, in which crustal thinning is accommodated by
the formation of normal faults and associated folds (Fig. 14-2;
also see Gibbs, 1983, 1984; Wernicke and Burchfiel, 1982).
This chapter focuses on cross sections of fold-thrust belts, for
most cross-section balancing studies to date have been applied
to these belts (e.g., Elliott and Johnson, 1980).

303

304

Special Topics Part II

In fold-thrust belts the package of rock above a fault is

called
a
thrust
sheet.
Thrus
Figure 14-2. Cross section illustrating geometry of folds
t
above a two-step normal fault. Note the anticline and syncline
sheet
Figure 14-1. Cross
section of a segment of the foothills belt of the Canadian Rockies that emphasizes the concept of thin-skinned s are
in the hanging-wall
block.
tectonics. The heavy lines are faults (arrows indicate relative movement; transport is toward the foreland). Faults do not name
penetrate below a detachment fault that lies in the plane of bedding near the base of the sedimentary sequence. (After Price, d for
14-2 1981.)
the
TER
fault that underlies them; for example, the body of rock lying
MINOLOGY OF FOLD-THRUST
above the McEachran thrust is called the "McEachran thrust
BELTS Structures of Fold-Thrust Belts
sheet" (Fig. 14-3a). The leading edge and trailing edge of a
thrust sheet are defined with respect to the transport direction
Fold-thrust belts develop either at convergent plate boundaries
of the thrust sheet. The transport direction is a vector in the
or as a result of continental collision, and they result in
map plane that defines the direction that the thrust sheet has
shortening or contraction of the crust. Next, we review the
moved at a given locality. For example, if a thrust sheet has
structural elements characteristic of fold-thrust belts.
moved to the east, the eastern edge is the leading edge, and the
(a) Internal and External Zones:
Broadly
western edge is the trailing edge. A forethrust is a thrust fault
speaking, orogenic belts in which fold-thrust belts occur can be
on which displacement is in the same general direction as
divided into internal zones and external zones. The internal
regional transport direction, and a backthrust is a thrust fault
zone is the portion of the belt in which plastic deformation
on which displacement is opposite to regional transport
dominates, penetrative strains develop, and metamorphism
direction. Similarly, the forelimb of an anticline is the limb
occurs., The internal zone is sometimes referred to as the
closer to the leading
hinterland. The external zone borders the undeformed
continental interior and is characterized by less plastic
deformation, nonmetamorphic conditions, and nonpenetrative
strains. The foreland of an orogenic belt refers, in a strict sense,
to the undeformed region in front of the thrust belt. Sometimes
this term is used with reference to the region of diminishing
shortening comprising the most external portion of the foldthrust belt.
(b) Detachments and Thrust Sheets: A detachment or
decollement is a subhorizontal or shallowly
dipping fault along which a sheet of rock has moved relative to
the underlying substrate. In a stratified sequence, detachments
commonly lie in the plane of bedding. Several detachments
may occur in a vertical sequence; in such a case the basal
detachment is the lowest one. The basal detachment can be a
regional fault that separates the entire package of rock
undergoing deformation and movement from the unaffected
rock below.
In external zones the basal detachment commonly forms
at or near the contact between sedimentary units and crystalline
basement (Dahlstrom, 1970; Fig. 14-1). In internal zones the
basal detachment commonly lies within crystalline basement
rocks. Therefore, the hanging wall of the fault in internal zones
contains basement (e.g., Harris, 1979; Cook et al., 1979;
Stanley and Ratcliffe, 1985). A detachment horizon or a glide
horizon is a stratigraphic interval in which detachments are
commonly found. In many cases detachment horizons are
composed of relatively weak rock, such as evaporite, but it is
dangerous to assume that simply because a unit is weak that it
must be a detachment horizon or that all detachments lie in
weak horizons. At some localities detachments run through
stiff units.
no vertical exaggeration

>
>
)

Chapter 14 Introduction to Cross-Section Balancing

305

r
of

edge of a thrust sheet, and the backlimb is the limb closer to

the trailing edge of the thrust sheet (Fig. 14-3).
(c) Ramp-Flat Geometry: To simplify thrust
geometries, it is common to portray thrust faults as smooth
planes in cross section. Many thrust faults actually have a
step-like profile (Fig. 14-3a) called ramp-flat geometry. A
thrust fault that cuts up to the syn-deformational erosion
surface is called an emergent thrust, whereas a thrust fault that
dies out in the subsurface is called a blind thrust (Fig. 3b). The
termination of a blind thrust at which displacement has
decreased to zero is called a tip line (Boyer and Elliott, 1982).
An exposed thrust fault may be an emergent thrust or simply a
blind thrust which has been exhumed by erosion.

Over part of its trace in profile a thrust fault lies in or

nearly in the plane of bedding and is parallel to bedding; these
segments of the fault are called flats. At other localities the
fault cuts more steeply across bedding; these segments of
faults are called ramps. In general, flats are much longer than
ramps. The intersection between a contact (e?g., a bedding
plane) and the fault is called a cutoff or cutoff line; hangingwall bedding is truncated by the fault at hanging-wall cutoffs,
and footwall bedding-plane contacts are truncated at fpotwall
cutoffs (Fig. 14-3a). The acute angle between the bed and the
fault at a cutoff is called a cutoff angle.
In practice it is necessary to distinguish between different
types of ramps and flats; Locations where the fault

T ransport

direction

l^t
Oil

nil

of

PC
ni

(a)
Emergent thrust

Blind thrust

r>
a
Jdt

Cb)

Figure 14-3.
(a) Cross section illustrating the step-like geometry of
a thrust fault. The definitions of a thrust sheet, leading edge, trailing edge,
backlimb, and forelimb are indicated. Footwall cutoff X was originally adjacent to hanging-wall
cutoff Y. Fault segment AB ^
_juxtaposesa footwall flat and a hanaina-wall flat. Fault

he
^ i (As. t

'*
IS

Ao
/5

juxtaposes a footwall ramp and a hanging-wall flat.

Fault segment
CD
juxtaposes a hanging-wall flat and a footwall flat.
Fault segment
DE
juxtaposes a hanging-wall ramp and a footwall flat. Fault segment EF juxtaposes a
hanging-wall flat against a footwall flat, (b) Cross section illustrating the difference
between an emergent thrust and a blind thrust (courtesy of D. Anastasio).

segment BC

306

Special Topics

is parallel to bedding of the hanging wall are called hangingwall flats, locations where the fault is parallel to bedding of
the footwall are called footwall flats, locations where the fault
cuts across bedding of the footwall are called footwall ramps,
and locations where the fault cuts across bedding of the
hanging wall are called hanging-wall ramps (Boyer and
Elliott, 1982; Woodward et al., 1985). After thrusting, a single
segment of fault can juxtapose either a hanging-wall ramp or a
hanging-wall flat against either a footwall ramp or a flat (Fig.
14-3a). For example, a single segment of a fault can be a ramp
with respect to the hanging wall and a flat with respect to the
footwall. We can describe a locality where such a
configuration occurs by saying that, "In this outcrop there is a
hanging-wall ramp on a footwall flat." A footwall flat is a
detachment
In general, the strike of a ramp is perpendicular to the
transport direction of the overlying thrust sheet. If the
orientation of a ramp with respect to regional transport
direction is such that the strike of the ramp is highly oblique
or even perpendicular to the transport direction (i.e., the fault
cuts up-section along strike), the ramp is called an oblique
ramp or a lateral ramp.
(d) Fault-Related Folds: Three major classes of folds
are associated with the development of ramp-flat fault
geometries. (1) The first class includes fault-bend folds
(Suppe, 1983). A fault bend is a change in dip of a fault
surface. Fault-bend folds develop in the hanging wall because
the hanging-wall block must bend to accommodate changes in
the shape of the fault (Fig. 14-3a). A hanging-wall anticline,
or ramp anticline, typically occurs above a ramp. Broad open
synclines, whose dimensions are controlled by the distance
between ramps, lie between
\,

Part II

(Dahlstrom, 1970; Jamison, 1987). These folds develop in

response to shortening above a detachment and are not
associated direcdy with ramps (Fig. 14-4b).
The geometry of folds exposed at the surface can be used to
predict fault geometry at depth. For example, in most (but not
all) localities, regions at the ground surface in which beds dip
away from the transport direction (e.g., transport was to the west
and the beds dip east) at the surface (region D; Fig. 14-5) reflect
areas where the upper block moved upward over a ramp in the
footwall. Forward bed dips (i.e., dips toward the direction of
transport) at the surface (region B; Fig. 14-5) occur above
hanging-wall ramp areas.
(e) Kink vs. Concentric Fold Styles: In many localities
folds that develop in fold-thrust belts are not smooth concentric
curves in profile but rather are subdivided into several dip
domains (Usdansky and Groshong, 1984; see also Faill, 1969) in
which the beds have a uniform dip (see Chapter 13). The dip
domains join with one another at an angular hinge (A, B, C, D,
and E are each distinct dip domains in Figure 14-5). In such
regions the fault-related folds are said to have a kink style. Kink
styles are associated with areas where the faults

adjacent ramp anticlines (Note: If ramps are very close, adjacent

ramp anticlines may merge). Fault-bend folds also form in
extensional terranes (Fig. 14-2), in which case they are
sometimes called rollover folds (see Gibbs, 1984; Hamblin,
1965). (2) The second class of folds includes
fault(a)
propagation folds (Suppe and Medwedeff,
1984;
Suppe, NNW
Mont Terri
anticline

1985).:
'These
folds
are the
result
of
flexural
bending of a layered sequence of rock in (b) advance of the
actual rupture and development of the fault plane (Fig. 14-4a).
(3) The third class of folds includes detachment folds

Clos du Doubs
anticline

SSE

Figure 14-4. Additional illustrations of

fault-related folds, (a) Fault-propagation
fold (adapted from Suppe, 1985); (b)
detachment folds above a subhorizontal
detachment in the Jura Mountains
(adapted from Laubscher, 1962).

>
I

Chapter 14 Introduction to Cross-Section Balancing

307

J
ft
m

i
*
In

he

Vi

111

re

U
ui

h
e.

themselves are planar between bends. There are localities,

however, where kink styles do not occur,
and concentric fold styles are more
appropriate (see Chapter 13).
It is easier to draw a cross section if
the folds are kink style, because limbs of
the folds can be drawn as straight-line
segments, and if there is no thinning or
thickening of beds, the distances between

from a basal detachment but do not rejoin at a higher

Figure 14-5. A single step in a thrust

fault and the associated anticline in the
hanging-wall thrust sheet illustrating the
terminology used for describing cutoff
angles and interlimb angles.

7= 60
d = 75
contacts can easily be kept constant. In addition, it $ -- <f> = 30 ft-- 60
stratigraphic
is easier to measure bed lengths and bed areas and to determine
level (Fig. 14-6a). (2) A duplex is a thrust system in which faults
cutoff angles on kink-style cross sections. For this reason most
cut up-section from a basal detachment and merge at a higher
of the cross sections drawn in this chapter are drawn with kinkstratigraphic level to form another continuous detachment (Fig.
style geometries.
14-6b). In a duplex the lower detachment is called the floor
(f) Fault-Bend Fold Angles:
Suppe (1983)
thrust, and the upper detachment is called the roof thrust. The
showed that the relationships between cutoff angles and faultfaults that cut up from the floor to the roof thrust surround
bend angles are not arbitrary. Specifically, for situations in
bodies of rock. These bodies, which are bounded on all sides by
which two flats connected by a ramp are parallel to one another
faults, are called horses. Duplexes occur in a range of scales.
(as in Fig. 14-5), bed length and layer thickness are conserved
Commonly, small duplexes form at the base of larger thrust
during faulting, and there is no slip between beds above flats,
sheets; in such a position, minor slip on each small fault in the
the relationship between fault-bend angles and cutoff angles can
duplex contributes to the overall movement of the thrust sheet
be described by a simple equation:
(Fig. 14-6c).
Duplex geometry in fold-thrust belts results in substantial
structural thickening in the vertical direction and substantial
<]> = 0 = tan-1 {sin 2y/(2 cos2 y + 1)} (Eq. 14-1),
shortening of the section in the horizontal direction. The
where 0 is the angle between the lower flat (the flat at the base
geometry of duplexes can be quite variable, depending on the
of the ramp) and the ramp, 4> is the angle between the upper
relative displacements on the faults within the duplex. Figure
flat (the flat at the top of the ramp) and the ramp, and 2y is the
14-7a shows an early stage in the evolution of a duplex in which
interlimb angle of the kinks above the hanging-wall cutoff (Fig.
the relative displacements on successive ramps are such that the
14-5). The angle between a ramp and a flat (0) is generally less
roof thrust is a smooth surface and is parallel to the floor thrust.
than 30 to 40, and for every 0 there are two possible values for
The final duplex composed of three horses is provided as Figure
y (called first mode and second mode). Usually, the shallower
14-7b. Figure 14-7c shows a duplex configuration in which the
limb dips (first-mode value for y) is observed. Additional
successive horses are stacked on top of one
trigonometric derivation (see Suppe, 1983) allows you to
calculate the hanging-wall cutoff angle (6), and the interlimb
angles (23) characterising the backlimb folds. It turns out, as an
example, that if 0 = 30, then B = 60; these angles used in
Figure 14-5.
The equation describing the angular relationship among
angles in fault-bend folds formed over faults in which the upper
and lower flats are not parallel is a bit more complex (see Suppe,
1983, 1985). For simplicity most of the cross sections described
in this chapter contain parallel upper and lower flats and obey
Equation 14-1. On kink-style cross sections, fault-bend angles
and cutoff angles can be easily displayed.
(g) Thrust Systems: A thrust system is an array of
kinematically related faults that developed in sequence during a
single regional deformation and are associated with deformation
above a basal detachment. There are two basic types of thrust
systems (see Boyer and Elliott, 1982, for more detail): (1) An
imbricate fan is a thrust system in which faults cut up-section

Figure 14-6. Illustrations, of thrust systems. (a) Imbricate fan;

(b) duplex structure (adapted from Perry, 1978); (c) small
duplex at the base of a larger one (adapted from Marshak,
1986).
Cb)

(adapted from Marshak, 1986).

another; such a configuration is called an antiformal stack. The

duplex of Figure 14-6b is one in which the roof thrust is not a
smooth surface that parallels the floor thrust; such a structure is
sometimes referred to as a lumpy-roofed duplex.
(h) Internal Strain in Thrust Sheets: In some localities a
significant portion of the total strain in *fold-thrust belts is
accommodated by formation of structures within thrust sheets
(e.g., Reks and Gray, 1983). Several different types of structures
can be considered in this category: (1) Backthrusts commonly
develop above hanging-wall anticlines (e.g., the east-verging
thrust faults in Figure 14-7c, which occur in a west-verging
fold-thrust belt). (2) Out-of-the-syncline thrusts (Fig. 14-8) are
faults that die out toward the hinge of a syncline, thereby
allowing rock to squeeze out of the core area of the syncline
when a room problem develops (i.e., there is insufficient space
for rock). Out-of-the-syncline faults can be either backthrusts or
forethrusts. These faults are sometimes called accommodation
structures. (3) Minor faults are faults on which displacement is
about an order of magnitude less than the displacement of
principal faults in the thrust system (Price, 1967; Wojtal, 1986).
(4) Minor folds are folds with amplitudes that are significantly
Figure 14-7. Types of duplexes, (a) Early stage
in the evolution of a duplex in which the roof
thrust ends up being parallel to the floor thrust.
The dashed line is an incipient thrust (adapted
from Boyer, 1978); (b) later stage in the
development of a flat-roofed duplex. Each fault
block confined between an upper and lower
thrust is called a horse. The thickness of the
lines represents the proportion of slip that is
transferred along any individual segment of fault.
In this figure the total slip on the upper
detachment to the east of the duplex equals the
slip on the lower detachment to the west of the
duplex, but slip on individual ramps in the duplex
is only a fraction of the total (adapted from
Boyer, 1978); (c) an antiformal stack of horses

smaller than the height of the thrust sheet (e.g., the folds in
thrust sheet A in Figure 14-7c).
In fold-thrust belts significant strain may also occur by
development of cleavage and/or by plastic shape change of
grains in the rock. The cleavage formed in fold-thrust belts is
typically spaced cleavage or slaty cleavage, the formation of
which involves pressure-solution deformation that may result in
volume-loss strain (e.g., Marshak and Engelder, 1985). Cleavage
is usually not uniformly distributed in fold-thrust belts (e.g., the
cleavage in Figure 14-7c is concentrated at the leading edge of
thrust sheet B).

Reference Lines
Three lines are commonly used to provide a reference frame for
describing relative movement and shortening in fold-thrust belts.

Chapter 14 Introduction to Cross-Section Balancing

Figure 14-8. Out-of-the-syncline thrust faults. Both a backthrust

and a forethrust are shown.

(a) Regional Dip and Regional Level: Prior to

displacement on faults and development of folds, strata in the
external zone of a developing orogenic belt are subhorizontal or
only shallowly dipping. The dip displayed by a package of strata
prior to folding and faulting is called the regional dip.
Generally, regional dips do not exceed a few degrees, and the
original depositional thickness of strata commonly increases in
the direction of regional dip. Folding and faulting not only result
in deviation from regional dip (e.g., if the regional dip is 1.5,
and the limb of the fold has a dip of 35, there is clearly a
deviation from regional dip) but also cause beds to be lifted
above a reference plane; this reference plane (a line in cross
section), which marks the elevation of a bedding surface before
deformation, can be called the regional level. We will see that
one of the steps involved in cross-section balancing requires
removing the effects of deformation so that beds "return to
regional dip and regional level." In this chapter, we use the term
structural relief to refer to the difference in elevation between a
surface (e.g., a bedding plane) at the top of a structure and the
same surface at regional level.
(b) Loose Lines:
It is important during the
restoration of a cross section to keep track of how much slip
between points in originally adjacent beds has developed in
different parts of the cross section. This is done by inserting a
loose line, which is merely a reference line drawn at an angle to
bedding either in the deformed or undeformed cross sections
(Fig. 14-5); a loose line does not extend below the basal
detachment and is usually placed near the trailing edge of the
cross section. A loose line can be an arbitrary line drawn
perpendicular to bedding, it can be the cross-sectional trace of a
fault, it can be a vertical line drawn at the trailing edge of a cross
section, or it can follow a known vertical drill hole.
A loose line can be considered to be a chain of marker
points in the layers of a sequence. If the loose line is marked on
a cross section of deformed rocks, then we can observe how it is
distorted or how it changes orientation during restoration of the
section. Inversely, if the loose line is inserted on the restored
section, we can observe how it distorts during our hypothesized
development of the deformation. The distortion or tilting of the
loose lines during these operations gives an indication of the
angular shear of beds past one another in different parts of a
structure. This shear is called interbed slip and is a consequence
of both shear on detachments and shear related to flexural
folding.
(c) Pin Lines: A pin line is another type of reference line

309

in a cross section. When we measure bed lengths, we need to

decide on some locality in the cross section at which to begin
our measurement. A pin line is merely the reference line at
which we begin measurement of bed lengths (Fig. 14-5). If our
section extends across the boundary between the defprmed belt
and an adjacent undeformed foreland, we choose a regional pin
line to lie in the undeformed foreland; a regional pin line can
extend all the way down to basement and can penetrate the cross
section either from the top or from the bottom.
Many cross sections, however, do not include undeformed
rock. In such cases an arbitrary local pin line must be chosen. It
is best to draw the local pin line perpendicular to bedding in the
least deformed part of a thrust sheet (where there has been no
interbed slip), where, in addition, there is the most complete
stratigraphic section. Therefore, local pin lines are commonly
drawn along a long flat, in a broad syncline, or at a fold hinge; a
local pin line cannot extend below the detachment at the base of
the thrust sheet containing the pin line. Pin lines should be
placed along a fold hinge only if rock-fabric data (e.g., the
occurrence of an axial-planar cleavage) indicates that the hinge
has not migrated (moved along the fault plane).

133 CONCEPT OF A BALANCED

CROSS SECTION
Types of Cross Sections
A new terminology has recently developed for discussing cross
sections; next, we introduce the names that are used to refer to
different types of cross sections (see also Elliott memorial
volume of the Journal of Structural Geology, 1983, v. 5, n. 2).
(a) Deformed-State Cross Section: A cross section that
represents the geometry of structures as they appear today, after
deformation, is called a deformed-state cross section.
(b) Admissible Cross Section: A deformed- state cross
section that depicts an interpretation in which structures look
like those that can be directly observed in mountain sides and
road cuts (i.e., the cross section depicts realistic-looking fold
and fault geometries) is called an admissible deformed-state
cross section. Whether or not a specific structure should be
deemed admissible depends not only on whether creation of the
structure is physically possible but also on whether such a
structure is likely to occur in a specific type of deformed belt.
Dahlstrom (1969, p. 743) pointed out that, "in a specific
geological environment, there is only a limited suite of
structures which can exist . . . For example, you should be
suspicious of a cross section that portrays a large recumbent
isoclinal fold with an extremely thickened hinge at the foreland
edge of a fold-thrust belt; such a structure would more likely be
found id the hinterland.
A restored cross section that depicts admissible structures
is called an admissible restored cross section. By saying that a
restored cross section is admissible, we mean that the fault
trajectories (the traces of faults on the cross section) shown on
the cross section are possible. Specifically, the angles between

310

ramps and flats should not be in excess of about 35.

(c) Restored Cross Section: A cross section that has been
"pulled apart," such that the fault displacement has been
removed and folds have been straightened out, is called a
restored cross section. The relative positions of rocks shown on
a restored cross section should be the relative positions that the
rocks had prior to deformation. Traces indicating the position
and attitude of surfaces that later became faults are also
typically shown on a restored cross section.
(d) Viable Cross Section: If a deformed-state cross
section can be restored to an unstrained state such that the
predeformation geometry of faults is admissible, bed lengths are
conserved and/or bed area is conserved, and bed lengths are
consistent, then the deformed-state cross section is said to be
viable. An understanding of what is meant by bed
"conservation" and "consistent" in this context is critical, so the
next part of this chapter is devoted to explaining the application
of these terms. Suppe (1985) uses the term retrodeformable
with much the same meaning as viable.
(e) Balanced Cross Section: A balanced cross section is a
deformed-state cross section that is both admissible and viable.
In other words, a balanced cross section portrays an admissible
suite of structures and can be restored such that the restored
cross section depicts consistency of bed length, conservation of
bed length and/or area, and admissible premovement fault
geometries. Additional constraints, described later in the
chapter, must also be met in order for a cross section to be
balanced.
The difference between a balanced cross section and an
"unbalanced" admissible cross section is that the balanced
version has been restored and tested for viability. Thus, when
someone says that they have "balanced" a cross section, they
mean that they have gone to the effort of restoring the section
and have tested the restored version.
It is important to keep in mind that by balancing a secdon,
you are not checking to see if it is correct; a balanced cross
section is still merely an interpretation that is quite possibly
incorrect. When new data become available, it is likely that the
section will need modification. A balanced section, however, is
at least possibly correct, whereas a section that does not balance
is probably wrong.

Considerations Involved in Testing the

Viability of a Restored Cross Section
/
In this portion of the chapter we clarify the concepts and
requirements for determining whether or not a cross section is
viable.
(a) Conservation of Area: If the deformation of a bed or
thrust sheet involves only folding and faulting the volume of a
bed or thrust sheet will not change during deformation. If, in
addition, the deformation yields plane strain, then the crosssectional area of the thrust sheet or bed does not change during
deformation. Figure 14-9a shows an undeformed layer of rock
containing a ramp on which there has not yet been displacement.
When the thrust sheet moves west over the ramp (Fig. 14-9b),

Special Topics

Part II

its cross-sectional area does not change and is still equal to that
shown in Figure 14-9a. In Figure 14-9c the shape of the ramp
anticline is different; the thrust sheet illustrated in this figure has
a different area than does the original undeformed thrust sheet,
and thus this figure illustrates a case where area has not been
conserved.
If the area of the thrust sheet in Figure 14-9a equals the
area of the sheet in Figure 14-9b, then the excess area (area
above regional level), which is labeled Ax, must equal Aj. Aj
does not equal Ax in Figure 14-9c. Figure 14-9d shows the same
fold form as does Figure 14-9c, but the trailing-edge of the
thrust sheet has been tilted sufficiently to make Aj equal to Ax
and thereby result in conservation of area.
Area conservation cannot be assumed for deformation that
involves development of volume-loss strain, such as commonly
accompanies the development of spaced cleavage involving
pressure solution. As noted earlier, significant pressure-solution
cleavage occurs locally in fold-thrust belts, so the measured
cross-sectional area of a deformed-state thrust sheet is locally
less than the original area. Volume-loss strain is usually not a
problem in the external portions of fold-thrust belts.
(b) Area Balance: If area conservation is assumed, the
area of a bed or a thrust sheet depicted on the deformed-state
cross section must equal the area of the thrust sheet as depicted
on the restored cross section. In other words, the restored area of
a thrust sheet must "balance" or correspond to the deformedstate area of the thrust sheet. In this chapter we call the operation
of comparing deformed-state and restored areas area balancing.
The area of a bed or thrust sheet can be measured by
subdividing it into simple geometric forms whose areas can

Chapter 14 Introduction to Cross-Section Balancing

b c

311

i"

' i

Cc)

d, = d2 A Aj a e t a
_UI

DC

s
Cd)

d, * d2 Ax s Aj , a'"e* ae

A, . d,

Figure 14-9. Illustration of the concept of bedlength and area conservation and balancing, (a)
Undeformed thrust sheet; (b) thrust sheet after
moving over a ramp. Area balance and bedlength balance are evident, and there is not
constant slip along the fault;
(c) cross section in which area balance and bedlength balance are not evident; (d) cross sectioh
in which there is bed-length balance and area
balance and constant slip along the fault. Points
a, b, c, d, and e are reference points, Ax is the
excess area, and Aj is the area between the pin
line and the trailing edge of the fault.

be calculated using plane geometry, by using a planimeter, by

digitizing, by plotting on graph paper and counting squares, or
by cutting out thrust sheet cross sections and weighing them.
(b) Conservation of Bed Length:
Conservation
of bed length refers to the supposition that the length of a
contact in cross section does not change during deformation.
Refer again to Figure 14-9a. Reference points along the top
surface of the bed are labeled a through e; the pin line is fixed
with respect to the footwall, to provide a reference frame. The
length of this contact in Figure 14-9b (line a'e', as measured
around the fold) equals the original length of ae. Therefore, bed
length was conserved during development of this ramp
anticline. The length of contact a"e" in Figure 14-9c is not equal
to ae, so bed length was not conserved during development of
this structure, * The length of contact a'"e'" in Figure 14-9d is
equal to ae, so bed length was conserved during development of
this structure.
(c) Bed-Length Balance: If bed length is conserved
during deformation, the length Of a contact (e.g., the top surface
of a horse) is the same in both the deformed-state and restored
cross sections. Such correspondence is called bed-length
balance. Whether or not bed lengths in the deformed-state and
the restored cross sections should balance depends on whether or
not the bed changed thickness during deformation and on
whether or not the strain was volume constant. We will see that
thickening of a layer during deformation means that bed-length
balance cannot occur if there is to be conservation of area.
(d) Consistency of Bed Length:
The total

lengths of each layer in a sequence depicted in a restored cross

section should be nearly the same or should vary in a consistent
manner. In other words, a straight loose line drawn perpendicular
to bedding at the trailing edge of a deformed-state cross section
should be either straight and perpendicular to bedding in the
restored section or smoothly varying in the restored section. Note
that if a vertical loose line were drawn at the trailing edge of the
thrust sheet in Figure 14-9d, it would be inclined to the west in a
restored version of this cross section.
Bed-length consistency can be determined by looking at the
restored shape of a loose line. It is important to emphasize that
consistency does not require that the restored loose line be
exactly perpendicular to bedding in the restored and deformedstate cross sections. At present the characteristics of acceptable
shapes for restored loose lines are not fully understood; it is fair
to say, however, that a sudden discontinuity or zigzag in a
restored loose line (Fig. 14-10) should be viewed as indicating
that bed lengths are not consistent. Bed-length consistency is
also indicated by restored fault geometry, as described next.
(J) Admissible Restored Fault Shapes: In the deformedstate cross section, fault geometries are distorted as a
consequence of movement on younger faults. In restored cross
sections, faults are depicted with the original shape that they had
prior to movement on younger faults. Therefore, faults in the
restored section should have reasonable step-like geometries
(such as is shown in Fig.
133). The abundant literature on fold-thrust belts suggests
that when a step-like fault initially forms, (a) ramp angles are
generally less than 35 with respect to bedding, (b) faults do not
turn back on themselves (i.e., segments of the fault trace do not
dip in opposite directions), and (c) faults do not cut downsection.

314

assumption simplifies the process of placing cutoffs on a cross

section as you draw it.
Constant slip on a fault, however, is not necessary and in
many cases is not even realistic. Good reasons for a change in
slip magnitude along the trace include (a) the occurrence of
simple shear along bedding planes in the thrust sheet (Fig. 1414; note that in this example slip on the detachment increases in
the direction of transport and that a loose line at the trailing
edge of the thrust sheet is inclined in the direction of transport),
(b) transfer of slip from another fault into the fault of interest at
their mutual intersection (Fig. 14-7b), and (c) the partitioning of
strain between fault displacement and other structures such as
folds and cleavage.

Figure 14-14. Changes in slip magnitude along the trace of the

fault resulting from shear along bedding planes in the thrust
sheet. Note that displacement on the upper .flat is greater than
displacement on the lower flat. (Adapted from Elliott, 1976.)

The last statement in the preceding paragraph requires

clarification. Look once again at Figure 14-9b. Note that dj >
d2, where d^ is the displacement of the trailing edge of the
thrust sheet with respect to the pin line, and d2 is the
displacement of the leading edge of the thrust sheet along the
upper detachment. The reason for the difference between dj and
d2 is that part of the shortening of the thrust sheet was
accommodated by formation of the fold. In the case of a faultpropagation fold (Fig. 14-4a), it is particularly clear that
displacement progressively decreases until it becomes zero at
the tip of the fault.

134 DRAWING A DEFORMED-STATE

CROSS SECTION
Next we illustrate the construction of a simple deformed- state
cross section. This example differs from those described in
Chapter 13 in that we demonstrate how to incorporate a fault
that is not exposed at the ground surface.
Problem 14-1
Given the strip map of a portion of a fold-thrust belt (Fig.
14-15a), construct an admissible deformed-state cross section
that accommodates the map data. Assume that the ground
surface is a horizontal plane. The only unit exposed at the
ground surface is unit h.
Method 14-1
Step 1: Examine the strip map (Fig. 14-15a). We know that
the map portrays a portion of a fold-thrust belt, so, although
there are no faults mapped, we suspect that there may be a

Special Topics

Part II

detachment at depth and that the fold portrayed may be related

to a ramp cutting up-section from the detachment.
Step 2: Determine the direction of transport. We see that
there are numerous strike and dip measurements. We take the
transport direction to be perpendicular to the regional strike.
Therefore, we choose an east-west line (XX1) to be our line of
section. Usually, we have enough information from regional
mapping to know which way the transport vector points. In this
example, the vector points west.
Step 3: Draw the topographic profile along the line of
section and plot the surface structural data (Fig. 14-15b). Draw
the stratigraphic column (presumed to be known from other
work) along the leading edge of the cross section to give an
indication of the position of contacts in the subsurface. We know
the depth of the top of unit g from the well data, and we know
that there must be enough room between the ground surface and
the basement to fit in the entire stratigraphic section of the
region. In fold-thrust belts, faulting thickens the section. Thus,
the undeformed thickness of strata in the area gives a minimum
depth to basement; basement may be deeper than this minimum
but not shallower.
Step 4: There are distinct dip domains, and thus we decide
to depict the anticline using the kink style of folding. We sketch
in unit g to illustrate the form of the anticline (Fig. 14-15b).
Notice that the top of unit g is at the same level on both sides of
the anticline, so we assume that the top of unit g represents a
regional level and that its dip is regional dip (horizontal). Using
the methods described in Chapter 13, we project the axial-plane
traces of kinks down to depth. We assume that there is no
thinning of strata on the limbs of the folds, so the traces of the
axial planes bisect the angle between the limbs of the folds (Fig.
14-15c).
Step 5: To complete the deformed-state cross section, we
propose a fault at depth. We begin with the hypothesis that the
fold at the ground surface is a ramp anticline. There is no way to
know where the fault is at depth based on the information
available. Constraints on the level of detachment may be
provided by knowledge of regional structural geology. Let's say
that we believe that the detachment lies at the base of unit d.
(a) To construct the ramp we work from east to west (in the
direction of transport). The easternmost kink hinge SS must
bisect the base of the ramp (point A), and the

>
>
D

Chapter
14 Introduction to Cross-Section Balancing
Dashed lines are dip-domain boundaries

315

t
i

wen

-VA -

I V V,,

y\
--V \

I
Cb)

/**

to

tv

at

is

11

Figure 14-15. Map and sections for problem 14-1. (a) Strip
map of a portion of a fold-thrust belt showing the attitudes
of bedding and the dip-domain boundaries; (b) cross
section showing dips in the plane of the section and the
regional stratigraphy to scale; (c) cross section showing
hypothetical fault; (d - f) alternative completed cross
sections that fit the same original data.

ramp must be parallel to the backlimb dip (Fig. 14-15c). We

draw the ramp cutting up-section to where it intersects the
second kink hinge TT' (point R). In this interpretation we
propose that the fault bends at point R and becomes a footwall
flat. Point R happens to fall on the unit d/unit e contact (this
coincidence is circumstantial and reflects the geometry of this
example).
(b) Now that we have drawn in the fault, we draw in the
footwall strata (Fig. 14-15c). We assume that the footwall strata
51
have regional dip.
(c) All that remains now is for us to draw in the core of
the anticline. We extend the kink hinges down to where they

intersect the fault surface and then draw in the formation

contacts so that they have the correct dips in the different dip
domains (Fig. 14-15d).
As we noted above, the interpretation shown in Figure
1315d is not a unique interpretation. Alternative cross
sections that fit the data in Figure 14-15a are shown in Figures
14-15e and 14-15f.

316

Special Topics

Considering that there are several possible solutions to

the data set, you may ask whether there are any definite
constraints on the geometry of a ramp at depth. The answer is,
fortunately, yes. Look at Figure 14-16; regardless of the depth
of the lower flat, the eastern synclinal axial-plane trace (KK')
must intersect the lower fault bend (point A in Figures 14-16a
and 14-16b). The eastern anticlinal axial-plane trace (JM) can
intersect the ramp at any point along the ramp until it reaches
the upper fault bend (point B in Figures 14-16a and 14-16b). If
it intersects at the upper fault bend (case 1), the western
anticlinal hinge (LM) and the western synclinal hinge must
intersect the upper flat at point B or to the west (Fig.
1316a). If JM intersects the ramp at some arbitrary
point along the ramp between points A and B (case 2), then
LM must intersect the upper fault bend at B (Fig. 14-16b).
Clearly, the ramp height (the vertical distance between the
lower and upper flats) and the displacement along the fault

Part II

a direct consequence of the evolution of a simple fault-bend

fold as described by Suppe (1983). Figure 14-17 illustrates
three stages in the development of a ramp anticline. Note that
the eastern synclinal hinge (ES) is fixed at the base of the ramp,
and the eastern anticlinal hinge (EA) migrates up the ramp with
increasing displacement. As long as EA intersects the ramp
below the upper fault bend, the western anticlinal hinge (WA)
is fixed at the upper fault bend. While EA migrates up the
ramp, WS moves west along the upper flat. When EA
intersects the upper fault bend, it becomes fixed at that
position, the distance between WA and WS is locked, and WA
and WS move to the west along the upper flat Notice that the
structural relief progressively increases as EA

Structur
al relief

Figure 14-16. Illustration showing the range of possible ramp

geometries that could fit the data of Figure 14-15a. (a) Fault
bends intersect axial-surface traces KK' and JM; (b) fault
bends intersect axial-surface traces KK' and LM.

are quite different depending on which interpretation (Fig.

1316a vs. Fig. 14-16b) is chosen, even though the lower
detachment is at the same depth in both cases. The maximum
possible depth for the fault is the depth at which point B
becomes point M. If the detachment is at this depth, case 1
becomes identical with case 2, and only one ramp height can
occur (shown in Figure 14-15e).
The relationships described in the previous paragraph are

Figure 14-17. Evolution of a fault-bend fold emphasizing the

migration and locking of axial trace positions, (a) EA is migrating
up the ramp, and WA is fixed at the upper fault bend; (b) EA and
WA both intersect the upper fault bend; (c) EA is locked at the
upper fault bend, and WA moves along the upper flat.

moves up the ramp. Once EA is at the top of the ramp, the

structural relief remains constant even as displacement on the
Chapter 14 Introduction to Cross-Section Balancing
fault increases and the distance between the hanging-wall
cutoff (C) and the footwall cutoff (U) increases.

13-

5 RESTORING A CROSS SECTION

In this portion of the chapter we describe the procedure of

restoring a cross section. Recendy, techniques have been
developed that allow this procedure to be done interactively
with a computer (e.g., Kligfield et al., 1986), but computer
procedures cannot be understood unless you are first adept at
restoring a section by hand. If bedding thickness does not
appear to have been changed during deformation (i.e., bedding
thickness is constant around folds), then restoration can be
carried out by assuming bed-length balance. If bed-thickness
change accompanying deformation is apparent in the
deformedUstatecross section, then bed- length balance cannot
be assumed and you must measure areas of thm^ sheets
When balancing a cross section, it is important that you
take into account the sequence of faulting; the youngest fault
should be restored first, and the oldest fault should be restored
last. This means that you should remove the effect of the
displacement on a younger fault before you remove the effect
of the displacement on an older fault. Consideration of the
sequence of faulting forces you to determine whether you
understand the evolution of structures portrayed in a cross
section.

Restoration Based on BedLength Balance

We start by restoring one of the deformed-state cross sections
produced in Method 14-1, then we will deal with more
complicated examples. In these problems bedding thickness is
constant around folds. Therefore, we can assume that area
balance occurs if bed-length balance occurs, and we can
restore sections merely by stretching out the contacts to return
them to regional level and regional dip. This method is
sometimes called the sinuous-bed method (Woodward et al.,
1985). We do not need to compare areas of beds or thrust
sheets. Measurement of bed lengths on the cross-section sketch
can be done direcdy with a ruler or dividers (Fig. 14-18) if the
folds are kink style. If the folds are concentric style, you may
need to use a planimeter, a waxed thread, or a computer
digitizer.
Problem 14-2
Restore the cross section shown in Figure 14-15d. Assume
that unit thickness is constant in the section, and assume that
the displacement is constant along the fault.

Dividers

317

CA' B') = CAB)

Cb)
Figure 14-18. Sketch showing how to use
dividers to measure line segments composing a
kink-style fold, (a) Measurement of line segment
AB on the fold; (b) restored bed. The length of
A'B' equals the length of AB.

Method 14-2
Step 1: Draft your cross section carefully; label the points
of intersection between the axial-plane traces and the bed
traces (Fig. 14-19a). To save space, units a and b are not
shown. The distance AB represents the distance between the
hanging-wall cutoff and the footwall cutoff on the unit d/unit e
contact and thus is the separation on the upper flat.
Step 2: Add reference lines to the cross section (Fig.
1319a). The cross section portrays only a portion of a
deformed belt; the detachment extends all the way to the west
edge of the figure, so there is no regional pin line. Draw a local
pin line perpendicular to bedding at the leading edge of the
cross section. The pin line does not extend below the fault, and
because it is only a local pin line, it moves during the
restoration. Draw a loose line (dotted line) at the trailing edge
of the cross section. The loose line does not extend across the
detachment and also moves during restoration.
Step 3: In a space below the deformed-state cross section,
draw a set of lines at appropriate spacing to represent the
stratigraphic sequence in the undeformed state. We call this set
of lines a stratigraphic frame. If the stratigraphic thickness
does not vary across the length of the cross-section area, the
lines are parallel (Fig. 14-19b). The lines should be longer than
the deformed-state cross section to accommodate the restored
lengths of the contacts.

B'

-0

---------------------------------------------------------------------------------------------------------

318
Step 4: Note that the detachment is not folded, so the
position of the ramp is fixed with respect to the footwall beds.
Therefore, we draw the ramp on the stratigraphic frame in the
same
orient W
ation
and

position with respect to the footwall Ca) beds as

depicted on the deformed-state cross
section.
Step 5: With

it

was

a pair of dividers,
measure
the
separation on the
upper flat (AB).
Point
A
coincides
with
the hanging-wall
cutoff at point 2,
and point B
coincides
with
the
footwall
cutoff at point 3. On the stratigraphic frame measure a distance
equal to AB starting at point 1 and extending east along the
unit d/unit e contact. Mark point r. Point 1' represents the
restored position of point
1. On the unit d/unit e contact line of the stratigraphic frame,
lay off the distance 1 to 2. Note that the restored position of
point 2 (i.e., point 2') lies directly over point
2. These points must coincide because when the
displacement on the detachment is removed, the hanging- wall
and footwall cutoffs must be juxtaposed. Now, lay off the
distances 2 to 4,4 to 5, 5 to 6, and 6 to 7 along the unit d/unit e
contact line to locate the restored positions 2', 4', 5', 6', and 7'.
The restoration of the unit d/unit e contact is complete (Fig.
14-19b).
Step 6: Repeat the procedure for the other contacts, and
locate the restored positions of points 8 thru 25. Restore from
the leading edge to the trailing edge, and start with the lower
contacts and move up-section. Draw the restored pin line and
loose line. Note that both of these reference lines moved east in
this example, because both lay above the detachment. Also
note that the horizontal distance between the loose line and the
pin line is greater in the restored cross section, because
restoration removed the folding that resulted from movement
on the detachment.
If the original thickness of units was not constant across
the cross section, the stratigraphic frame could not consist of
parallel lines. If strata thicken continuously in a given

Special Topics

Part II

direction, the stratigraphic from should depict a thickening

wedge. The exact rate of thickening cannot be specified until
the wedge is actually restored. But if you can estimate the
approximate amount of shortening depicted on the
E
deformed-state cross section, you can draw an
approximate stratigraphic frame by depicting layers that
change progressively from the thickness shown at the
leading edge of the cross section to that shown at the
trailing edge of the cross section (Fig. 14-20).

Figure 14-19. Cross sections for problem 14-2. (a)

The deformed-state cross section with the reference
lines shown; (b) the stratigraphic frame and the
completed restored cross section.

>

IF

>
|

Chapter 14

Introduction to

Cross-Section Balancing

Figure 14-20. Thickening stratigraphic wedge. Layers thicken

from east to west, so AB > CD. Restored faults are shown.
Note that the thickness of units in thrust sheet 1 is greater
than the thickness in thrust sheet 2.

Problem 14-3
Restore the deformed-state cross section shown in Figure
14-21a.
Method 14-3
Step 1: Scan the cross section. It shows three thrust faults
that emerge from a basal detachment. Two of the faults bound
a horse, and two of the faults are truncated at the ground
surface by erosion. The strata below fault A appear to be in
place. Thus, a pin line that penetrates these strata at the west
edge of the cross section can be considered to be a regional pin
line that is fixed during the restoration. Draw the regional pin
line and a loose line (Fig. 14-21a).
Step 2: Bedding thickness is constant in the deformedstate cross section, so below the cross section we create a
stratigraphic frame by drawing a series of appropriately spaced
parallel lines (Fig. 14-21b).
Step 3: Because bed lengths are constant, we can restore
by measuring bed lengths (assuming that there has not been
significant volume-loss strain). Begin measuring at the regional
pin line and work to the east. Start with the beds at the base of
the section. Fault A, the youngest fault, plots on Figure 14-21b
in the same position and orientation as it did in the deformedstate cross section. The restored positions of points 1 through 5

sheet A that lie above the shaded bed; because these contacts
are truncated by erosion, we do not know the entire length of
319our
these contacts in thrust sheet A. In order to continue
restoration, therefore, we must introduce a local pin line in
sheet A. This pin line is drawn perpendicular to bedding at the
point where the stratigraphic sequence is most complete and is
presumably little deformed (Fig. 14-21 a).
Step 5: Now we can continue the section restoration. On
the partially restored section, draw the local pin line
perpendicular to bedding (Fig. 14-21b). The restored position
of the local pin line can be determined from where it cuts the
shaded bed. The depiction of the restored pin line as a straight
line reflects our assumption that the beds have not slipped past
one another in the vicinity of the pin line. Measure bed lengths
between the pin line and fault B in the deformed-state cross
section, and locate the restored

Local pi n l i ne

cross section; (b) stratigraphic template for the restoration, and partial restoration; (c) completed restoration.

along the top of the shaded bed are shown. Point 5 shows the
restored position of the footwall cutoff of the top contact of the
shaded bed.
Step 4: A problem arises when we try to restore contacts in

320

Special Topics

position of fault B. Measure bed lengths to the west of the local

pin line to determine the position of the trace of the erosional
truncation surface on the restored section (Fig.
14-21b).
Step 6: Repeat the procedure to restore the remaining
thrust sheets, and find the restored position of the loose line
(Fig. 14-21c). We discuss problems with this restored Cross
section later in the chapter.

Restoration Based on Area Balance

In general, if a bed is significantly thickened during deformation, we cannot simply stretch out the bed to restore the
cross section; rather, we must restore the cross section such
that the areas of restored and deformed-state beds are the
same.
Problem 14-4
Consider the rather unusual-Iooking ramp anticline shown
in Figure 14-22a. It is obvious that there has been extreme
thickening in the hinge. The area of unit b in the thrust sheet
shOwn in this figure is 322 units1. Determine the restored
length of unit b in the thrust sheet.
Method 14-4
Step 1: First we must create a stratigraphic frame. The
trailing edge of the thrust sheet contains unthickened unit b
that is 4 units thick. Therefore, we take the undeformed
thickness of unit b to be 4 units and create a stratigraphic
frame on which unit b has this thickness (Fig.
14-22b).
Step 2: Examine the cross section (Fig. 14-22a) and prove
that restoration based on bed-length balance will not work.
Contact MN on the deformed-state cross section is 50 units
long. If we unfold contact MN and place it on the stratigraphic
frame, we should get a restored bed that is 50 units long and 4
units thick, with an area of only 200 units 2. The folded bed,
however, has an area of 322 units2, so this restored bed has an
area that is 38% too small. Thus, reconstruction by such a
procedure would yield a restored cross section that does not
meet the constraint of area conservation.
Step 3: To restore unit b of the thrust sheet we measure its
area (322 units2) and determine a bed length on the
stratigraphic frame that is long enough to yield this area.
Assuming that the original bed thickness is 4 units (as it is
under point N), the restored bed should actually be 80.5 units
long (Fig. 14-22c). The area of the shaded layer on this figure
is also 322 units2.

146 EVALUATING AND IMPROVING A

CROSS SECTION
Checking the Balance of a Cross Section
In this portion of the chapter we provide examples of how to
check whether a cross section balances and if it does not, how
to alter it to achieve balance.

(C)

Part II

Problem 14-5
Determine if the cross section used in problem 14-3 (Fig.
14-2la) is balanced.

10
________________J

UNITS

N'
Au : 200

(to

Figure 14-22. Cross sections for problem 14-4.

(a) Deformed
state cross section; (b) restored bed created by simply stretching out the
top contact of the shaded layer. The area of this bed is too small; (c)
restored bed created by assuming conservation of area.

321

Chapter 14 Introduction to Cross-Section Balancing

Method 14-5
Step 1: First, we look at the deformed-state cross section
(Fig. 14-2la) and check to see that the structures shown are
admissible. Since there are no impossible fault shapes, we
accept this cross section as admissible. Now we look to see if
the restored cross section is viable.
Step 2: Determine if bed lengths are consistent. To do this
we look at the shape of the deformed-state loose line (Fig. 1421c). It curves back on itself (reverses dip). This shape is
worrisome and suggests that the deformed- state cross section
may not be viable.
Step 3: Check the shape of the restored fault traces (Fig.
14-21c). All of them have admissible step-like patterns and
admissible ramp angles.
Step 4: We conclude, based on the observation that the
restored loose line may have an unacceptable shape, that there
is a problem with the viability of the restored cross section, and
we do not yet accept the original deformed-state cross section
as balanced.
Problem 14-6
Determine if the deformed-state cross section provided in
Figure 14-23a is balanced.
Method 14-6
Step 1: Examine the deformed-state cross section (Fig. 1423a). The numbered points are reference points on the cross
section, and the patterned bed is a marker bed. At first glance it
looks admissible.
Step 2: Look to see if the template constraint is obeyed. It
is not. We note, for example, that in thrust sheet A, the length
of the ramp that begins at point 12 and cuts up-section across
the unit is not matched by the length of the corresponding
hanging-wall ramp, even after the difference between the
hanging-wall and footwall cutoff angles is accommodated.
There are several comparable problems in this section.
Step 3: Construct a restored cross section (Fig.
14-23b).
Step 4: Examine the restored cross section (Fig.
1423b). Note that the restored loose line looks
unreasonable and that the restored faults reverse dip. From
these geometries we conclude that the
deformed-state cross section is not balanced.
By going through the exercise of restoring the
Figure 14-23. Cross sections for problem
14-6. (a) Deformed- state cross section;
(b) restored
section, our attention
cross section.

has been drawn to

problems in the section that we might otherwise have missed.

322

Improving a Cross Section

At this point you are probably asking the most important
question of all, namely, What can be done if a cross section
does not balance? How can a cross section be corrected or
improved? The most common errors that experienced
geologists make in constructing a cross section include ( 1)
depicting an incorrect depth to detachment, and (2) depicting
insufficient shortening at a given level in the cross section. We
will show how to deal with these problems in the following
examples.
Problem 14-7
In Problem 14-5 we faced a situation where the restored
loose line reversed dip at the base of the restored cross section.
Change the deformed-state cross section of Figure 14-21a so
that this problem is removed.
Method 14-7
Step 1: Think about the significance of the problem. We
conclude that the base of the patterned bed is too short. By
referring back to the deformed-state cross section, we notice
that if we change the dip of the lower part of fault
C, we can increase the length of the shaded bed in thrust sheet
B.
Step 2: Make the necessary change (Figure 14-24a). Check
the bed-length balance of the new deformed-state cross section
by restoring it (Fig. 14-24b). The restored loose line does not
reverse dip and thus looks reasonable.

Ca
)

Chapter 14 Introduction to Cross-Section Balancing

323

Figure 14-24. Cross sections for

problem 14-7. (a) Deformed- state cross
section, showing correction; (b) restored
version of the modified deformed-state
cross section. Note that the restored
loose line does not reverse dip.

Thus, we can say that the restored cross section is viable and
thus that the deformed-state cross section is balanced.
The correction described above is not the only possible
change that could improve the cross section. For example,
rather than increasing the length of beds lower in the section,
we could shorten beds higher in the section. A shortening of
the beds could be accomplished by changing the hanging-wall
cutoff angles.
Problem 14-8
Figure 14-25a shows a deformed-state cross section. This
cross section is not admissible. Improve the deformed-state
cross section.
Method 14-8
Step 1: Study the cross section (Fig. 14-25a). We
immediately see some potential problems. First, the
displacement on the thrust drastically increases up-dip; the top
of unit b is only slightly offset, whereas the top of unit c is
offset considerably. Also, the template constraint is not met;
the hanging-wall cutoffs are not matched by the footwall
cutoffs.
x<
it 1
Y.

Step 2: Redraw the

section
with
an
appropriate change. We
relocate the lower flat to
I.

CaJ

Cb)

Figure 14-25. Cross sections for problem 14-8.

Inadmissible cross section; (b)
possible cross section.

(a)

ensure that the template constraint is met (Fig. 14-25b). This

cross section is admissible. You may prove to yourself that it is
i

also viable.
Problem 14-9
Figure 14-26a provides dip data along an east-west line of
section. Assume that the line of section is parallel to the
transport direction. A well drilled along the line of section
provides stratigraphic data at depth. Draw a balanced cross
section of this area. This example demonstrates that in practice,
you should keep in mind the potential viability of the restored
cross section while you draw the deformed-state cross section.
Method 14-9
Step 1: Examine the dip data and stratigraphic data
provided in Figure 14-26a. Notice that it is possible to
recognize dip domains. Also notice that stratigraphic section
seems to be repeated at depth. It appears that the well
penetrates a location where a hanging-wall flat lies over a
footwall flat.
Step 2: We use the stratigraphic data from the well to
define unit thickness. With this information plus the dip data
we project kink-style folds down to the depth of the base of
unit a. Below the well we can draw in the repeated stratigraphy
(Fig. 14-26b). Examine the resulting cross section. It appears
that the base of unit a is indeed a detachment. The problem
remains, How do we fill in the space in the cross section below
the detachment?
Step 3: We speculate that along the western edge of the
cross section the thrust sheet has been emplaced over flat-lying
strata. In Figure I4:26c we draw in the stratigraphic section and
extend the contacts to the east. The geometry looks viable
below the syncline and below the westernmost ramp. But we
still have a bothersome gap to fill between the anticline
(indicated by stipple in Fig.
1426c). Clearly, a solution in which we have only one
ramp originating from the basal detachment does not work. We
speculate that the space can be filled by inserting a

I
3
#

f
324

Special Topics

Part II

IS

le

la

11

Figure 14-26. Maps and cross sections

for problem 14-9.
(a)
Dip-domain data, positions at which
contacts cross the ground surface, and a
well showing stratigraphy at depth (a - d
are stratigraphic units) provided along a
section line; (b) partial construction of a
deformed- state section, showing folded
layers and a folded thrust. Question
marks indicate space that must be filled;
(c) partial solution showing how the
space at the west edge of the section
can be filled with flat-lying strata. Stipple
indicates area that still needs to be filled;
(d) solution showing a horse filling the
space below the anticline; (e) partially
restored cross section showing the
restored shape of the horse.

(c)

(d)

Ce
)

horse, so we sketch in a possible ramp that could form the base

of the horse (heavy dashed line in Fig. 14-26c). This line is
drawn to be parallel to the backlimb of the anticline.
Step 4: We complete the deformed-state cross section by
completing the ramp at the base of the horse and by filling in
the strata within the horse (Fig. 14-26d). The deformed-state
cross section looks admissible.
Step 5: We restore the cross section. Figure 14-26e
provides a partial restoration that shows the geometry of the
region before movement on fault A occurred. Notice that the
anticline disappears, and the trailing edge of the horse becomes
a simple ramp; the restored cross section looks viable. Based
on this partially restored section, we feel that the deformedstate cross section of Figure 14-26d is probably balanced. We
do not have the data to restore the major thrust sheet, because
we have no exposures of hanging-wall cutoffs and therefore do
not know the displacement on the fault. We do recognize,
however, that the displacement is substantial, because section
is doubled across the entire length of the cross section.
Notice that in Method 14-8 we used the word speculate
several times. That is because the process of constructing the
deformed-state cross section shown in Figure 14-26d is an
intuitive process guided by our prior knowledge of geometries
in fold-thrust belts. Our end product looks reasonable, but we
have no way of confirming that the cross section is correct
unless we can obtain sufficient drill-hole or seismic data. Note
how the procedures of cross-section balancing have made us
think about every line drawn in the section.
One final comment about correcting cross sections: In
some regions a duplex forms at depth and results in significant
shortening of the lower part of the stratigraphic sequence.
Units above the roof thrust do not appear to show evidence of
comparable shortening. For example, the restored length of
unit Dc in Figure 14-27a is much less than the restored length
of Dm (Fig. 14-27b). This discrepancy reflects the fact that in
the deformed-state cross section (Fig. 14-27a) Dm is
duplicated by faulting, whereas Dc is only folded. There are

real examples of formations where the shortening of the

"folded-only" strata above the roof thrust is much less than the
shortening of a fault-imbricated sequence below the roof
thrust. This discrepancy can be explained in a number of ways,
one of which is shown in Figure 14-27c; in this cross section,
shortening of unit Dm is shown to involve not only the large
folds but also mesoscopic folding and cleavage

Chapter 14 Introduction to Cross-Section Balancing

325

Regional pm lino /Tip line

Ca)

1 Missmg area
i**

**

Figure 14-27. Alternative solutions to the problem of discrepancy

in the amount of shortening between strata above and below a
roof thrust, (a) Deformed-state cross section. Regional pin line is

shown to the west of the tip line; (b)

restoration, showing the difference
in bed length between the, supraand subroof thrust strata; (c)
discrepancy explained by the
formation of volume-loss cleavage.
(Adapted from Geiser, in press.)

formation, which did not appear in cross section 14-26a. This

extra shortening makes up the missing area. For discussion of
additional explanations, see Geiser (in press).

147 DEPTH-TO-DETACHMENT AND

REGIONAL SHORTENING CALCULATIONS
The principle of area conservation that we described earlier
can be applied to calculate the depth to the detachment beneath
a detachment fold. This calculation works only if there are no
additional thrusts between the folded surface and the
detachment.
Problem 14-10
A shallow cross section of a folded surface is provided in
Figure 14-28a. Determine the depth to the detachment below
this structure, assuming that it is a detachment fold.

I.

Method 14-10
Step 1: Consider the folded surface between points A and
G to be a marker line (Fig. 14-28a). The length of this folded
line is Lu, which represents the undeformed length of this bed
(the distance between points A and G before deformation).
Trace the folded marker line onto a new piece of paper.
Step 2: Connect two points along the line by a straight
chord (the dashed line between A and G in Fig.
1428a). This chord represents regional level and
should be parallel to the detachment. The length of the chord is
Ld (and represents the distance between A and G in the
deformed state). The difference between Lu and Ld represents
the amount of shortening of the bed that resulted from
formation of the fold. Measure L and Ld. L,j = AB + BC + CD
+ DE + EF + FG = 39 units, and Ld = AG measured along the
chord = 32 units. Therefore, L,j - Ld = s = 7 units.
Step 3: The area between the trace of the folded surface
and the dashed line between points A and G is called

(b)

_unils_J ^
Figure 14-28. Cross sections for problem 14-10
(depth-to-detachment calculation), (a) A folded
marker horizon. Excess area is shaded; (b)
construction for depth-to-detachment calculation; (c)
cross section showing properly located detachment;
(d) alternative solution showing two imbricate horses.
Note that in this solution the depth to the detachment
is much less.

i
>

'wmm & w

Chapter 14 Introduction to Cross-Section Balancing

the excess area (Ax). We measure the area by counting

squares on a grid placed under the figure and determine that
Ax = 101 units2.
Step 4: Extend chord AG to point H; the position of point
H is determined so that the length of straight line AH is equal
to Lu. If there is conservation of area, the area (Aj) between
the line segment GH and the detachment at depth d must also
equal the excess area, Ax. This relationship indicates that if the
detachment remains at a given depth, shortening of the bed
resulted in movement of a point from G to H and creation of
the excess area Ax above chord AG. Therefore,
s=

Ly - L,j

Ax = Aj = (s)(d)
d = Ax/s
(Eq. 14-2).
Step 5: Using this equation, we can calculate the value of
d (14.4 units), which is the depth to the detachment. Note that
all we needed to measure was the lengths Ly and Ld and the
area Ax under the fold. All these values can be obtained
directly from the cross section.
Step 6:
Redraft the cross section showing the
detachment at depth (Fig. 14-28c).
Note that in the solution shown in Figure 14-27c we filled
the space between the folded marker bed and the detachment
with folded rock. The cores of the anticline could contain
ductile evaporite or shale that squeezed upward during
shortening or a layered sequence that is folded
disharmonically with respect to the overlying strata. We
cannot know for sure without additional data.
If, alternatively, imbrication on thrust faults had occurred
between the detachment and the folded marker bed, our
calculation of depth to detachment would be incorrect. In
Figure 14-28d we show the same folded marker layer, but now
we show it to be above the roof thrust of a duplex. Note that
with this geometry the depth to detachment is much less. In
other words, a depth-to- detachment calculation works only if
there are no faults between the folded surface and the
detachment. Furthermore, note that in the duplex
interpretation of Figure
1428d the bed above the roof thrust slipped to the left
out of the cross section.
In cases where it is not possible to calculate the depth to
detachment, this depth may be predicted from knowledge of
regional stratigraphy. Also, the age of the oldest rocks brought
up as a hanging-wall flat is a clue to the identity of the unit in
which the detachment lies.
If faults are present beneath a fold, the excess area above
regional level can still indicate the amount of shortening
resulting from movement of the thrust sheet (Laubscher, 1962;
Gwinn, 1970). Consider Figure 14-9b as an example. If the
fault geometry a n d depth to detachment are known, and area
has been conserved, then the excess area (A x), which can be
measured on the cross section, determines the value of Aj, and
the amount of shortening (dj) is merely Aj/h, where h is the
depth to the deeper detachment.

326

148 APPLICATIONS OF BALANCED

CROSS SECTIONS
As we have emphasized already, the principal use of balancing
procedures is in providing constraints on cross sections that
must be constructed from incomplete data. Because balanced
cross sections are more likely to be correct, they are obviously
useful in guiding resource exploration. The application of
cross-section balancing techniques to petroleum exploration is
evident in many papers, such as Bally, Gordy, and Stewart
(1966), Dahlstrom (1969, 1970), Royse, Warner, and Reese
(1975), and Lamerson (1982).
Balanced cross sections are also useful in regional
tectonic analysis, because they provide more reliable
constraints on the production of palinspastic maps. A
palinspastic map is a map that shows the distribution of
stratigraphic units prior to their deformation. Such maps are
constructed by removing the effects of fault displacement and
folding (see Kay, 1945; Dennison and Woodward, 1963;
Dennison, 1968). A palinspastic map gives a much clearer
image of the spatial distribution of stratigraphic units and/or of
early structures that formed prior to the restoration of the
structures. For example, a palinspastic map of the Valley and
Ridge fold-thrust belt of the Appalachians gives a much clearer
image of the dimensions of the Paleozoic sedimentary basins
that existed along the eastern margin of North America prior to
the occurrence of convergent tectonics (e.g., collisional events)
in the region. A palinspastic map of the Cenozoic Basin and
Range province in the North American Cordillera can be used
to show the spatial relationships of major Mesozoic and
Paleozic deformed belts and depositional belts of the
Cordillera. The restored cross section that is constructed during
cross-section restoration provides data on the positions of units
and markers before deformation and thus allows construction
of a palinspastic map.

14-

9 ACKNOWLEDGMENTS

The reviews of David Anastasio, Sandy Figures, Scott

Wilkerson, and Steven Wojtal greatly improved this chapter.
We appreciate additional comments from Tim Byrne, Peter
Geiser, and Steven Usdansky.

Chapter 14 Introduction to Cross-Section Balancing

327

EXERCISES

1.

Complete the deformed-state cross section of the Bristol Peak region shown in Figure
14-MI. Assume that bed thickness is Constant across the map area, that the
Precambrian/Cambrian contact is not folded, and that the ground-surface trace gives the
shape of a simple ramp anticline.

2.

Figure 14-M2 shows a simple cross section of a ramp anticline above the Mashpee
thrust.
(a) Label the hanging-wall and footwall ramps and flats on this cross section. Does
this cross section obey the template constraint?
(b) If not, construct a cross section that displays the same stratigraphy and the same
fault geometry and obeys the template constraint.

Figure 14-M1. Cross-sectional sketch of the Bristol Peak area for exercise 14-1. (Adapted from Woodward et al., 1985.)

Mashpee thrust
Figure 14-M2. Cross-sectional sketch of the Mashpee thrust for exercise
14-2.

3 . The geologist working on a cross section of the Alice Hills region has been called off on
another project. You are left to develop a cross-sectional interpretation of the region.
Your predecessor has left several useful hints to make the task easier; you have access
to a partially completed cross section (Figure 14-M3). Note that the region displays
kink-style folds and that the dip-domain boundaries are already labeled. The dip angles
in the nonhorizontal domains are indicated. Note that there are well data at three

328

Special Topics

localities. If you measure (Jlosely, you will see that the Silurian layer in hole 3 is almost
twice as thick as in the other two holes. The upper interval of Silurian strata is not as
thick as the lower interval. Finally, studies elsewhere in the region suggest that beds
have uniform thickness in the region and that detachment horizons occur along
formation contacts.
(a) Complete the cross section by showing the structure between the ground surface
and the Precambrian/Cambrian contact. You must choose detachment levels, position
ramps, and indicate the amount of slip on each ramp.
(b) Restore your completed section. Does it balance?
3.

Figure 14-M4 shows a cross section of the Appalachian Mountains fold-thrust belt
extending across Virginia and West Virginia from the Blue Ridge out onto the
Appalachian Plateau. The original version of this cross section is by T. H. Wilson and
appears in Woodward (1985). The real geology has been modified a little in drawing
Figure 14-M4 in order to make this exercise easier to work.
(a) Describe the type(s) of thrust system(s) that are displayed on this cross section.
(b) In comparison with the structural geometries shown in the chapter, are the
structures displayed on the cross section admissible?
(c) What is the direction of transport? Where do the principal detachments occur?
(d) Number the thrust sheets that contain the Cambro-Ordovician carbonate strata.
Assuming a break-forward sequence, in what order were the thrust sheets emplaced?
(e) Assume that the leading edge of the cross section is a regional pin line and stays
fixed during deformation. (This assumption is flawed, because the cross section
indicates that the upper flat extends at least to the west edge of the cross section).
Restore the cross section. If you wish, you may assume constant bed thickness to
simplify the restoration, even though it is clear that the Cambro- Ordovician carbonate
sequence thickens to the southeast. You may restore assuming bed-length balance.
(0 Is the deformed-state cross section balanced?
(g) Using your restoration, compare the shortening of the strata above the roof thrust
with the shortening of the strata below the roof thrust. Discuss the discrepancy.
Remember that the pin line you used is probably not a regional pin line.
(h) Calculate the percent shortening of the Cambro-Ordovician sequence by
comparing your restored and deformed-state cross sections. If you wish, repeat the
calculation using the excess-area method.

5 . Figure 14-M5 provides a cross section of the Appalachian fold-thrust belt in the vicinity of
the Powell Valley anticline, Tennessee.
(a) Briefly describe the thrust system depicted in this cross section.
(b) Restore this cross section and check the balance. (The geology has been modified
slighdy to facilitate this exercise). Be sure to choose a regional pin line.
(c) Calculate the shortening of the patterned layer in the Knox Group.

Part II

CO

ro

oo

J,r,, k,rn ,,,

Alice Hills
Figure 14-M3. Structural data along a line of section in the Alice Hills
region for exercise 14-3.

ZT
0)
a

<0

NW

Figure 14-M4. Cross section of the Appalachian Valley and Ridge Province for exercise 14-4. (adapted from Woodward, 1985.)

CO

ro

CO

Figure 14-M5. Cross section of the Powell Valley Anticline region for exercise 14-5. (adapted from Suppe, 1985.)

Chapter 14 Introduction to Cross-Section Balancing

I
>
I

6 . Construct two restored versions of Figure 14-M6 (Kingston Ridge region), one using bed-

length balance and the other using area balance. Compare the two and explain which
you think is the more appropriate technique for restoring this particular cross section.

Kingston Ridge region

Figure 14-M6. Cross section of the Kingston Ridge region for exercise 14-6.

7.

Restore the cross section of Figure 14-M7 (Ben Jesse region). Note that the stratigraphy
in the section increases in thickness toward the trailing edge of the thrust system. You
will need to insert local pin lines in order to complete the restoration. Be sure to show
the position of the erosion surface on the restored version.

Ben Jesse region

Figure 14-M7. Cross section of the Ben Jesse region for exercise 14-7 (adapted
from Woodward et al., 1985).

8.

Calculate the depth to detachment beneath the top of the stippled layer in Figure
14- 4b using the method described in Section 14-9. Is the detachment level shown in Figure 144b correct?

9.

Once the exercises in this chapter have been completed, we recommend that students be
challenged to produce balanced cross sections of real map areas. Due to

332

333

Special Topics

Part II

cost, we could not provide real maps with this book. Maps that provide appropriate data for cross-section
balancing exercises include U.S.G.S. quadrangle maps 1-686 (Afton 30' quadrangle) and 1-1129 (Cokeville
30' quadrangle), which are maps of the Idaho-Wyoming fold-thrust belt by W.W. Rubey, and the 1:125,000scale maps of the Tennessee portion of the Valley and Ridge fold-thrust belt by Rodgers (1953; Tennessee
Division of Geology, volume 58, part II). Cross sections by Roeder et al. (1978), Woodward (1985), and
Suppe (1980) provide a variety of cross sections at a useful scale for study.