Modelling Landscape Evolution - Environmental Modelling: Finding Simplicity in Complexity - page 311

kinematics and homogeneous bedrock lithologies char-
acterizing most rifted margins lend themselves well for
first-order numerical modelling. Early conceptual models
(Kooi and Beaumont, 1994; Tucker and Slingerland,
1994) studied under what conditions escarpments could
be maintained under geologically long periods of time.
These models showed that escarpments can be inherited
features from the pre-rift topography without requiring
syn- or post-break-up surface uplift, and pointed
toward two fundamentally different types of escarpment
evolution, through parallel retreat (as envisaged in most
of the classical literature on the subject) or through
rapid down-wearing of the region seaward of the
escarpment (Figure 19.6). Subsequent models developed
to simulate the evolution of the SE Australian and SE
African margins (van der Beek and Braun, 1999; van der
Beek et al ., 2002) suggested that surface uplift at both
margins dates from before continental break-up and
that both margins evolved in the plateau downwearing
style. Although both styles of evolution lead to very
similar present-day topography and drainage patterns
(Figure 19.6), the denudation history of the margin as
recorded by low-temperature thermochronology data
permits to discriminate between them (van der Beek
et al ., 2002; Braun and van der Beek, 2004).

it. In both these models, tectonic advection took place
vertically only, whereas most orogenic belts record large
lateral displacements of rocks. Willett et al . (2001) were
the first to study the effects of such lateral advection on
orogenic topography (Figure 19.7), exploiting the capac-
ity of the Cascade code to include lateral motion, and
showed that (1) the observed topographic asymmetry of
many mountain belts is directly controlled by horizontal
advection and (2) true topographic steady state is unlikely
to be reached in orogens undergoing lateral advection.
Herman and Braun (2006) applied a similar model (but
with more elaborate kinematics) to the Southern Alps of
New Zealand and showed that, indeed, strong asymme-
tries in tectonic uplift and tectonic advection, together
with rapid alteration between glacial and interglacial con-
ditions, constantly interact to prevent the landscape from
reaching topographic steady state.

19.7 Conclusions and outlook

The above review, although incomplete, has been
intended to show in what way landscape-evolution
models have contributed to our understanding of how
topography evolves through time and what controls
this evolution. At the present state of development,
most landscape-evolution models remain of limited use
as predictive tools, primarily due to our incomplete
understanding of the processes controlling landscape
development, the non-uniqueness of model predictions,
and the problems of up- and down-scaling associated with
modelling surface processes (see also Chapter 5). Rather,
the main value of these models is heuristic: they permit
hypotheses to be explored and quantify potential con-
sequences of inferred evolutionary scenarios (Bras et al .,
2003). These restrictions are not unique to landscape-
evolution models, but are more generally inherent to
modelling in the Earth Sciences (Oreskes et al ., 1994).
Even though the application of landscape-evolution
models to diverse geomorphic problems has increased
rapidly in the last two decades, some scepticism remains as
to their usefulness when applied to problems on geological
space- and timescales: the models may be perceived as
too simplistic, both in their implementation of process
laws and in their boundary conditions. Nevertheless,
such models appear to be an appropriate tool to address
first-order questions as: 'How do tectonics and erosional
processes interact in the formation and evolution of
topography? How does climate change affect erosion and

19.6.4 Orogenicbelts

A very large body of recent work has focused on poten-
tial couplings and interactions between tectonics, climate
and surface processes in active orogens (see review by
Whipple, 2009). Given this widespread interest in the
topography of mountain belts, it is perhaps surprising
that only few authors have used landscape-evolution
models to study it. The reason for this relative paucity
probably lies in the fact that, in contrast to the rifted
margins discussed above, both the kinematics and the
variability of exposed bedrock are complex in orogenic
belts and modelling inherently requires strongly simpli-
fying the problem. As an example of such an approach,
K uhni and Pfiffner (2001) modelled the evolution of
drainage patterns in the European Alps in response to
spatial and temporal variations in uplift, but experienced
difficulties for their model to reproduce the significant
drainage diversions implied by the sedimentary record of
the Alps. Garcia-Castellanos et al . (2003) focused on the
Ebro foreland basin of the Pyrenees, rather than on the
mountain belt itself, to study the controls on the evolution
of the basin, in particular the causes of onset and termi-
nation of the major endorheic phase that characterizes