Geology Reference
In-Depth Information
magnitude of differential uplift and folding)
shows that the fold noses propagate and
amplify vertically at rates of about 1 km/Myr and
100 m/Myr, respectively (Bennett
et al
., 2000).
Although these rather slow rates imply that these
Otago folds are several million years old, the
prevalent semi-arid climate has limited overall
downwearing of the fold surface to about 30 m,
and drainages are fairly lightly etched into the
schist that underlies each fold.
When the drainages develop in folded ranges
with weakly lithified sediments draping over
bedrock (Fig. 10.15), another interesting
uncertainty emerges concerning the develop-
ment of the current fluvial network. Did today's
drainage patterns develop due to interactions of
rivers with the mantle of overlying sediments
or with the resistant bedrock once it became
exposed during deformation (Oberlander, 1985;
Tucker and Slingerland, 1996)? Were rivers that
pre-existed folding “let down” or superposed on
to the deforming landscape, such that they are
independent of the growing structures? Have
rivers that currently cut across folds shifted later-
ally during fold growth (Fig. 10.18) or did they
maintain an antecedent course? Commonly, the
answer to this must be gleaned from the strati-
graphic record where changes in provenance
and river systems can be documented. In
response to prolonged convergence, imbricated
folds and thrust faults build entire mountain
belts. The drainages that develop on these ranges
might be expected to differ in their geometric
properties in response to contrasts in climate,
lithology, rock uplift rate, and/or erosion rates
in different settings. In fact, such variation is
remarkably limited among many ranges. Instead,
studies of the spacing (
S
) of major drainages and
comparisons to the half-width (
W
) of the ranges
on which they developed demonstrate a strik-
ingly consistent relationship, with a spacing ratio
(
W
/
S
) of about 2. This consistency suggests that
some of the early-formed irregularities in river
courses that developed in response to individual
growing folds or faults are smoothed during the
drainage expansion and competition among
drainages that occur as mountains grow (Talling
et al
., 1997). This consistent ratio has another
intriguing implication: as a range widens, some
laterally adjacent drainages must be captured to
maintain the drainage spacing.
Past stream captures are typically difficult to
document with the preserved geomorphic record.
Wind gaps are defined as topographic saddles
through which a now-defeated river formerly
flowed. But, how do we distinguish such a saddle
from a pass on a ridge that never had a river
flow across it? Most typically, little evidence is
preserved of a former river. Some of the best
indicators of a pre-existing, through-flowing
river include the following: (i) preservation of
sediments that indicate long-distance fluvial
transport - well-rounded clasts, well-sorted sand-
stones, mature sediment shapes and compositions,
and non-local lithologies; (ii) fluvial terrace rem-
nants that are now folded across the water gap;
and (iii) remnant fluvial channels that do not scale
with the current catchment geometry - channel
widths or the depths of channel incision do not
decrease toward the current drainage divide. In
the Otago folds, for example, quite broad channels
etched into the bedrock can be traced through a
wind gap, up and over the fold crest.
True wind gaps where former rivers have been
defeated clearly indicate that some catchment
has gained drainage area and discharge at the
expense of another, that is, a drainage capture
has occurred. But which river and how do we
recognize it? In their study of the Otago folds,
Jackson
et al
. (1996) use both verifiable and
assumed wind gaps and the known direction of
fold propagation to make logical deductions
about how the drainage network progressively
evolved as the folds grew. Other geomorphic
criteria that can be used to deduce river capture
and reversal of drainage directions include the
following: (i) river terraces that tilt in the opposite
direction to the modern river gradient and are
unrelated to any local structural deformation;
(ii) “barbed” rivers whereby river junctions define
atypical angles, for example, by pointing upstream
or entering at nearly right angles even though the
converging rivers have similar discharge and low
gradients; (iii) provenance indicators that are
inconsistent with the current upstream source
area; and (iv) convoluted catchment geometries.
If the stratigraphic record of sediments
derived from the drainage can be examined,
