Introduction -
Fault growth is achieved by increases of both surface dimensions and displacement. These changes accrue in incremental slip events
that can be accompanied by fault linkage and slip localisation for both individual faults and fault systems. In the 1980s and 1990s
the majority of studies used mainly 2D map-view data in combination with finite fault dimensions and displacements to constrain fault
growth. A positive correlation between fault length and maximum displacement was interpreted to indicate that both increase progressively
during growth. In such a model increase in fault length is achieved by propagation of individual faults, with or without the linkage of
initially isolated faults. More recently, the increased availability of 3D seismic reflection datasets permits the use of growth strata
to chart the evolution of fault lengths and displacements throughout the history of faulting. These studies, however, provide few examples
of propagating faults and instead support a growth model in which fault displacements accrue progressively for the duration of faulting
and dimensions are achieved early in the history of faulting. Fault interactions are generally inferred to prevent, or at least retard, fault
propagation, with interaction between faults from an early stage of faulting being a key element of this growth model.
Fault growth models proposed in the literature differ primarily in terms of two basic criteria that define the extent to which individual
faults, or fault segments, were in kinematic ‘isolation’ from one another in their early stages of growth and to their ‘propagation’
histories. The fault isolation criterion defines a continuum from faults that remain isolated throughout their growth to faults that were
interacting and part of a coherent array from the onset of faulting. Few believe that faults remain isolated for the duration of their growth,
with isolated fault models generally requiring initial isolation followed by later interactions: differences in fault growth are therefore
defined along an isolationinteraction spectrum, which we discuss in the next section. The propagation criterion also defines a continuum, the
limits of which are characterised by models where the final length and maximum finite displacement are only achieved at the cessation
of faulting and, at the other extreme, where the final (maximum) length is reached geologically instantaneously at the start of faulting and
displacements accrue throughout deformation. As we will see later in this chapter, these phenomena are often related because the degree to
which faults interact generally has a first-order impact on their propagation. Determining where faults are on the isolation/interaction and
propagation spectrums can be challenging with the currently available datasets. It can, for example, be very difficult, if not impossible, to
discriminate between faults that interacted from their initiation and those that did so from an early stage in their faulting history, because
the required time-series information from growth strata are not available or poorly resolved.
Here we briefly present geometric and kinematic constraints for each of the two criteria, in an outline that partly reflects the progression
of understanding of fault growth over the past five decades. This outline will provide the essential backdrop to a more detailed consideration
of existing growth models for tectonic faults followed by their reconciliation with constraints derived from earthquake studies. As we progress
through our description, it will become clear that faults range in dimensions from millimetres to 100s of kilometres, often with fractal scaling
properties, and are components of complex systems. In these circumstances the distinction between individual faults, fault zones (discrete zones
in which displacement is distributed onto two or more slip-surfaces) and fault systems can be subjective and dependent on the scale of observation,
which is ultimately controlled by the methods of data acquisition. Despite the limitations imposed by the scale of observation, the nucleation,
propagation, interaction and linkage of individual faults can be interpreted with reference to the isolated/interaction and propagation criteria.
Since the most complete datasets constraining fault growth are available for normal faults, they are the main focus of this chapter, however,
data for reverse and strike slip faults suggest that the interpreted patterns of growth may be independent of fault-slip type.
In: Understanding Faults: Detecting, Dating, and Modeling. (Edited by Tanner, D. & Brandes, C.). Elsevier, ISBN: 9780128159859.