AT&G
- Spotila's active tectonics and geomorphology research
Controls on the long-term erosion of active mountain
belts
This is an erosion rate map of the San Bernardino Mountains,
inferred from thermochronometry and geologic data. Colors are coded for
erosion rate as shown. Construction of these erosion models for the San
Bernardino and San Gabriel Mountains is possible because of the wealth of data
constraining recent exhumation. These maps enable a comparison of
long-term erosion rate with boundary conditions, such as precipitation and
bedrock type. From this figure it is clear that the location of major
structures (e.g. the San Andreas fault along the southwestern margin or the
North Frontal thrust system along the northern margin of the range) play an
important role in erosion.
Scope: Most geomorphology focuses on erosion, as
it is the dominant process providing the first-order character to nearly all
landscapes. There are several ways of addressing how erosion works.
One avenue of research involves empirical and theoretical study of how specific
erosional agents or mechanisms (e.g. fluvial bedrock incision) behave.
Another synthesizes the mathematical laws of particular agents of erosion into
computer simulations of long-term landscape evolution. Both of these are
critical to understanding erosion, but missing from each is an empirical view
of how landscapes actually erode over the million year timescale. What
boundary conditions are most important? What controls long-term rates of
erosion? One way of investigating these questions is to compare the
erosion of young, active mountain ranges to various parameters that may
influence erosional processes.
In one study we have used the Transverse Ranges as a template
for a test of what controls long-term erosion of mountain belts. The San
Bernardino and San Gabriel Mountains share several similar boundary conditions
(similar size, relief, climate, structural setting, and both are nearly all
crystalline bedrock whose history shares at least some basic similarities) yet
have completely different geomorphic expressions. Is the greater
ruggedness of the San Gabriel Mountains due to greater erodibility? Is it
due to orographically-induced precipitation? Could it reflect a greater
topographic maturity and a longer history of active mountain building?
Because the erosional history of these two ranges has been constrained by
numerous thermochronometric studies and various geologic data, it is possible
to compare long-term erosion (exhumation, to be precise) rate with conditions
that may be important; such as bedrock type, precipitation magnitude,
structural position, etc. The figure above shows an erosion map for the
San Bernardino Mountains, that is the key interpretation needed for this
comparison. Results show that long-term erosion rate is correlated with
nearly all parameters that are thought to be important in controlling
erosion. Current thinking is that the fastest erosion of mountains occurs
where numerous independent parameters that foster erosion are coincident,
whereas protection from erosion, which can lead to plateau formation, occurs
only where numerous parameters that hinder erosion are coincident; a line of
thinking that can be called "coincident determinism." An
example of this occurs in the San Bernardino Mountains, where linear bodies of resistant
metamorphic rock occur as pendants within more erodable batholithic
rocks. The more resistant linear bodies helped to force asymmetric,
east-west drainage divides along the northern and southern margins of the Big
Bear plateau, thus forcing all denudation to proceed inwards from the plateau's
tapering flanks. The independent positioning of these metamorphic rocks,
and their coincident location with east-west thrust faults on the north and
south, has thus had an important role in protecting the plateau from erosion
and thus shaping the landscape. The importance of drainage divides in
controlling the long-term erosion of other mountain belts must be further
investigated, such as in the Tibetan Plateau. Another step that must be
taken is to quantify the importance of each parameter and thus go beyond this
somewhat anecdotal interpretation of how mountain belts erode.
Another approach to the boundary conditions that affect
erosion is to examine the effect of different variables on specific erosional
processes. For example, in our
study of headwater
channels in the Appalachians, variable bedrock strength was clearly a
dominant factor in shaping channels and controlling erosional process. In fact, it would not be possible to
predict what the topography of the Valley and Ridge would look like today,
without first knowing the precise depositional, burial, and deformation history
of sedimentary units whose variable resistance to erosion is responsible for
the shape of the landscape. This
has motivated a potential future area of research for my group, that focuses on
finding better ways to quantify rock resistance to erosion and its distribution
in a landscape in 4-d.
Personnel: Synthesis of the landscape evolution
of the San Bernardino and San Gabriel Mountains included collaborations with Martha House (Caltech), Ann Blythe (USC), Nathan Niemi
(Caltech), and Greg Bank. I also hope to collaborate with Rick Law (VPI) on
landscape evolution studies of the Mt. Everest massif to
address similar questions from slightly different perspectives. Rick has
an on-going project including collaborations with Mike Searle (Oxford) and Kip Hodges (MIT)
on the deformation history of the Greater Himalayan wedge between the Main
Central thrust and South Tibetan detachment system.
Follow
this link to more pictures of a deeply weathered surface in the San
Bernardino Mountains, that is important for constraining the magnitude of
recent exhumation and relating it to boundary conditions.
Mt. Everest at dawn, taken by Dr. Rick Law in May, 2000 from Tibet
(the Rongbuk Valley). This is
another location future research may involve, that focuses on the role of
boundary conditions in shaping topography and denudational patterns in a
landscape. What controls
erosion? It's a good place to
address this question.
Comments to: spotila@vt.edu